From 9b3814e2c551a9c9263a5f83632f98909b0e605a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E5=BE=90=E5=85=86=E6=A5=A0?= Date: Thu, 24 Jul 2025 13:12:44 +0000 Subject: [PATCH] feat: refactor model manager * chore: mv model icon * fix: model icon * fix: model icon * feat: refactor model manager * fix: model icon * fix: model icon * feat: refactor model manager See merge request: !905 --- .../api/handler/coze/workflow_service_test.go | 17 +- backend/application/application.go | 12 +- .../application/base/appinfra/app_infra.go | 7 + .../appinfra/modelmgr.go} | 118 +++-- .../appinfra/modelmgr_test.go} | 5 +- backend/application/modelmgr/init.go | 197 +------- backend/application/modelmgr/modelmgr.go | 62 +-- backend/application/singleagent/create.go | 33 +- backend/application/singleagent/get.go | 15 +- backend/application/singleagent/init.go | 5 +- backend/application/workflow/init.go | 2 +- backend/conf/model/entity/.gitkeep | 0 backend/conf/model/meta/.gitkeep | 0 .../entity/model_entity_template_ark.yaml | 94 ---- ...l_entity_template_ark_doubao-1.5-lite.yaml | 66 --- ...tity_template_ark_doubao-1.5-pro-256k.yaml | 66 --- ...ntity_template_ark_doubao-1.5-pro-32k.yaml | 66 --- ..._template_ark_doubao-1.5-thinking-pro.yaml | 90 ---- ...te_ark_doubao-1.5-thinking-vision-pro.yaml | 90 ---- ...y_template_ark_doubao-1.5-vision-lite.yaml | 90 ---- ...ty_template_ark_doubao-1.5-vision-pro.yaml | 90 ---- ...ty_template_ark_doubao-seed-1.6-flash.yaml | 90 ---- ...template_ark_doubao-seed-1.6-thinking.yaml | 90 ---- ...l_entity_template_ark_doubao-seed-1.6.yaml | 90 ---- ..._entity_template_ark_volc_deepseek-r1.yaml | 66 --- ..._entity_template_ark_volc_deepseek-v3.yaml | 66 --- .../entity/model_entity_template_basic.yaml | 132 ------ .../entity/model_entity_template_claude.yaml | 47 -- .../model_entity_template_deepseek.yaml | 72 --- .../entity/model_entity_template_gemini.yaml | 90 ---- .../entity/model_entity_template_ollama.yaml | 47 -- .../entity/model_entity_template_openai.yaml | 131 ------ .../entity/model_entity_template_qwen.yaml | 66 --- .../meta/model_meta_template_ark.yaml | 41 -- ...del_meta_template_ark_doubao-1.5-lite.yaml | 40 -- ...meta_template_ark_doubao-1.5-pro-256k.yaml | 40 -- ..._meta_template_ark_doubao-1.5-pro-32k.yaml | 40 -- ..._template_ark_doubao-1.5-thinking-pro.yaml | 41 -- ...te_ark_doubao-1.5-thinking-vision-pro.yaml | 42 -- ...a_template_ark_doubao-1.5-vision-lite.yaml | 41 -- ...ta_template_ark_doubao-1.5-vision-pro.yaml | 42 -- ...ta_template_ark_doubao-seed-1.6-flash.yaml | 42 -- ...template_ark_doubao-seed-1.6-thinking.yaml | 42 -- ...del_meta_template_ark_doubao-seed-1.6.yaml | 42 -- ...el_meta_template_ark_volc_deepseek-r1.yaml | 40 -- ...el_meta_template_ark_volc_deepseek-v3.yaml | 40 -- .../meta/model_meta_template_basic.yaml | 75 ---- .../meta/model_meta_template_claude.yaml | 40 -- .../meta/model_meta_template_deepseek.yaml | 35 -- .../meta/model_meta_template_gemini.yaml | 48 -- .../meta/model_meta_template_ollama.yaml | 31 -- .../meta/model_meta_template_openai.yaml | 39 -- .../meta/model_meta_template_qwen.yaml | 36 -- .../model/template/model_template_ark.yaml | 133 ++++++ .../model_template_ark_doubao-1.5-lite.yaml | 108 +++++ ...odel_template_ark_doubao-1.5-pro-256k.yaml | 108 +++++ ...model_template_ark_doubao-1.5-pro-32k.yaml | 108 +++++ ..._template_ark_doubao-1.5-thinking-pro.yaml | 131 ++++++ ...te_ark_doubao-1.5-thinking-vision-pro.yaml | 132 ++++++ ...l_template_ark_doubao-1.5-vision-lite.yaml | 131 ++++++ ...el_template_ark_doubao-1.5-vision-pro.yaml | 132 ++++++ ...el_template_ark_doubao-seed-1.6-flash.yaml | 132 ++++++ ...template_ark_doubao-seed-1.6-thinking.yaml | 132 ++++++ .../model_template_ark_doubao-seed-1.6.yaml | 132 ++++++ .../model_template_ark_volc_deepseek-r1.yaml | 108 +++++ .../model_template_ark_volc_deepseek-v3.yaml | 108 +++++ .../model/template/model_template_basic.yaml | 201 +++++++++ .../model/template/model_template_claude.yaml | 90 ++++ .../template/model_template_deepseek.yaml | 107 +++++ .../model/template/model_template_gemini.yaml | 139 ++++++ .../model/template/model_template_ollama.yaml | 84 ++++ .../model/template/model_template_openai.yaml | 171 +++++++ .../model/template/model_template_qwen.yaml | 106 +++++ .../contract/crossmodelmgr/cross_modelmgr.go | 39 -- backend/crossdomain/impl/modelmgr/modelmgr.go | 52 --- backend/crossdomain/workflow/model/model.go | 9 +- .../internal/agentflow/agent_flow_builder.go | 4 +- .../internal/agentflow/agent_flow_runner.go | 5 +- .../internal/agentflow/node_chat_model.go | 9 +- .../singleagent/service/single_agent_impl.go | 3 + backend/domain/modelmgr/entity/chat_model.go | 41 -- .../domain/modelmgr/entity/chat_model_test.go | 89 ---- .../domain/modelmgr/entity/model.schema.go | 23 - backend/domain/modelmgr/interface.go | 69 --- .../internal/dal/dao/chat_model_entity.go | 136 ------ .../internal/dal/dao/chat_model_meta.go | 189 -------- .../internal/dal/model/model_entity.gen.go | 31 -- .../internal/dal/model/model_meta.gen.go | 34 -- .../domain/modelmgr/internal/dal/query/gen.go | 111 ----- .../internal/dal/query/model_entity.gen.go | 417 ----------------- .../internal/dal/query/model_meta.gen.go | 425 ------------------ .../domain/modelmgr/service/model_manager.go | 389 ---------------- .../service/model_manager_with_cache.go | 41 -- .../workflow/crossdomain/model/model.go | 5 +- .../crossdomain/model/modelmock/model_mock.go | 29 +- .../internal/compose/test/llm_test.go | 2 +- .../workflow/internal/compose/to_node.go | 4 +- .../internal/nodes/llm/model_with_info.go | 15 +- .../workflow/internal/nodes/llm/prompt.go | 2 +- backend/infra/contract/chatmodel/config.go | 6 +- .../contract}/modelmgr/const.go | 18 +- .../contract/modelmgr/desc.go} | 42 +- backend/infra/contract/modelmgr/modelmgr.go | 27 ++ .../infra/impl/modelmgr/static/modelmgr.go | 80 ++++ .../agent/singleagent/model_mgr_mock.go | 54 ++- backend/internal/testutil/chat_model.go | 2 +- backend/types/ddl/gen_orm_query.go | 26 +- .../volumes/minio/default_icon}/claude_v2.png | Bin .../minio/default_icon}/deepseek_v2.png | Bin .../volumes/minio/default_icon}/doubao_v2.png | Bin .../volumes/minio/default_icon}/gemini_v2.png | Bin .../volumes/minio/default_icon}/ollama.png | Bin .../volumes/minio/default_icon}/openai_v2.png | Bin .../volumes/minio/default_icon}/qwen_v2.png | Bin 114 files changed, 2888 insertions(+), 4982 deletions(-) rename backend/application/{modelmgr/init_by_env.go => base/appinfra/modelmgr.go} (58%) rename backend/application/{modelmgr/init_by_env_test.go => base/appinfra/modelmgr_test.go} (83%) delete mode 100644 backend/conf/model/entity/.gitkeep delete mode 100644 backend/conf/model/meta/.gitkeep delete mode 100644 backend/conf/model/template/entity/model_entity_template_ark.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-lite.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-pro-256k.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-pro-32k.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-thinking-pro.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-thinking-vision-pro.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-vision-lite.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-vision-pro.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_ark_doubao-seed-1.6-flash.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_ark_doubao-seed-1.6-thinking.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_ark_doubao-seed-1.6.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_ark_volc_deepseek-r1.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_ark_volc_deepseek-v3.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_basic.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_claude.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_deepseek.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_gemini.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_ollama.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_openai.yaml delete mode 100644 backend/conf/model/template/entity/model_entity_template_qwen.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_ark.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-lite.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-pro-256k.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-pro-32k.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-thinking-pro.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-thinking-vision-pro.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-vision-lite.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-vision-pro.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_ark_doubao-seed-1.6-flash.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_ark_doubao-seed-1.6-thinking.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_ark_doubao-seed-1.6.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_ark_volc_deepseek-r1.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_ark_volc_deepseek-v3.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_basic.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_claude.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_deepseek.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_gemini.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_ollama.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_openai.yaml delete mode 100644 backend/conf/model/template/meta/model_meta_template_qwen.yaml create mode 100755 backend/conf/model/template/model_template_ark.yaml create mode 100755 backend/conf/model/template/model_template_ark_doubao-1.5-lite.yaml create mode 100755 backend/conf/model/template/model_template_ark_doubao-1.5-pro-256k.yaml create mode 100755 backend/conf/model/template/model_template_ark_doubao-1.5-pro-32k.yaml create mode 100755 backend/conf/model/template/model_template_ark_doubao-1.5-thinking-pro.yaml create mode 100755 backend/conf/model/template/model_template_ark_doubao-1.5-thinking-vision-pro.yaml create mode 100755 backend/conf/model/template/model_template_ark_doubao-1.5-vision-lite.yaml create mode 100755 backend/conf/model/template/model_template_ark_doubao-1.5-vision-pro.yaml create mode 100755 backend/conf/model/template/model_template_ark_doubao-seed-1.6-flash.yaml create mode 100755 backend/conf/model/template/model_template_ark_doubao-seed-1.6-thinking.yaml create mode 100755 backend/conf/model/template/model_template_ark_doubao-seed-1.6.yaml create mode 100755 backend/conf/model/template/model_template_ark_volc_deepseek-r1.yaml create mode 100755 backend/conf/model/template/model_template_ark_volc_deepseek-v3.yaml create mode 100755 backend/conf/model/template/model_template_basic.yaml create mode 100755 backend/conf/model/template/model_template_claude.yaml create mode 100755 backend/conf/model/template/model_template_deepseek.yaml create mode 100755 backend/conf/model/template/model_template_gemini.yaml create mode 100755 backend/conf/model/template/model_template_ollama.yaml create mode 100755 backend/conf/model/template/model_template_openai.yaml create mode 100755 backend/conf/model/template/model_template_qwen.yaml delete mode 100644 backend/crossdomain/contract/crossmodelmgr/cross_modelmgr.go delete mode 100644 backend/crossdomain/impl/modelmgr/modelmgr.go delete mode 100644 backend/domain/modelmgr/entity/chat_model.go delete mode 100644 backend/domain/modelmgr/entity/chat_model_test.go delete mode 100644 backend/domain/modelmgr/entity/model.schema.go delete mode 100644 backend/domain/modelmgr/interface.go delete mode 100644 backend/domain/modelmgr/internal/dal/dao/chat_model_entity.go delete mode 100644 backend/domain/modelmgr/internal/dal/dao/chat_model_meta.go delete mode 100644 backend/domain/modelmgr/internal/dal/model/model_entity.gen.go delete mode 100644 backend/domain/modelmgr/internal/dal/model/model_meta.gen.go delete mode 100644 backend/domain/modelmgr/internal/dal/query/gen.go delete mode 100644 backend/domain/modelmgr/internal/dal/query/model_entity.gen.go delete mode 100644 backend/domain/modelmgr/internal/dal/query/model_meta.gen.go delete mode 100644 backend/domain/modelmgr/service/model_manager.go delete mode 100644 backend/domain/modelmgr/service/model_manager_with_cache.go rename backend/{api/model/crossdomain => infra/contract}/modelmgr/const.go (69%) rename backend/{api/model/crossdomain/modelmgr/modelmgr.go => infra/contract/modelmgr/desc.go} (86%) create mode 100644 backend/infra/contract/modelmgr/modelmgr.go create mode 100644 backend/infra/impl/modelmgr/static/modelmgr.go rename {backend/conf/model/icon => docker/volumes/minio/default_icon}/claude_v2.png (100%) rename {backend/conf/model/icon => docker/volumes/minio/default_icon}/deepseek_v2.png (100%) rename {backend/conf/model/icon => docker/volumes/minio/default_icon}/doubao_v2.png (100%) rename {backend/conf/model/icon => docker/volumes/minio/default_icon}/gemini_v2.png (100%) rename {backend/conf/model/icon => docker/volumes/minio/default_icon}/ollama.png (100%) rename {backend/conf/model/icon => docker/volumes/minio/default_icon}/openai_v2.png (100%) rename {backend/conf/model/icon => docker/volumes/minio/default_icon}/qwen_v2.png (100%) diff --git a/backend/api/handler/coze/workflow_service_test.go b/backend/api/handler/coze/workflow_service_test.go index 147a6e5d..1836b404 100644 --- a/backend/api/handler/coze/workflow_service_test.go +++ b/backend/api/handler/coze/workflow_service_test.go @@ -18,7 +18,6 @@ package coze import ( "bytes" - "context" "errors" "fmt" @@ -49,7 +48,6 @@ import ( "gorm.io/gorm" modelknowledge "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/knowledge" - crossmodelmgr "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" plugin2 "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/plugin" pluginmodel "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/plugin" "github.com/coze-dev/coze-studio/backend/api/model/ocean/cloud/playground" @@ -85,6 +83,7 @@ import ( "github.com/coze-dev/coze-studio/backend/domain/workflow/entity" "github.com/coze-dev/coze-studio/backend/domain/workflow/entity/vo" "github.com/coze-dev/coze-studio/backend/domain/workflow/service" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" "github.com/coze-dev/coze-studio/backend/infra/impl/checkpoint" "github.com/coze-dev/coze-studio/backend/infra/impl/coderunner" mockCrossUser "github.com/coze-dev/coze-studio/backend/internal/mock/crossdomain/crossuser" @@ -1503,7 +1502,7 @@ func TestNestedSubWorkflowWithInterrupt(t *testing.T) { }, } - r.modelManage.EXPECT().GetModel(gomock.Any(), gomock.Any()).DoAndReturn(func(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *crossmodelmgr.Model, error) { + r.modelManage.EXPECT().GetModel(gomock.Any(), gomock.Any()).DoAndReturn(func(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *modelmgr.Model, error) { if params.ModelType == 1737521813 { return chatModel1, nil, nil } else { @@ -1972,7 +1971,7 @@ func TestReturnDirectlyStreamableTool(t *testing.T) { }, } - r.modelManage.EXPECT().GetModel(gomock.Any(), gomock.Any()).DoAndReturn(func(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *crossmodelmgr.Model, error) { + r.modelManage.EXPECT().GetModel(gomock.Any(), gomock.Any()).DoAndReturn(func(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *modelmgr.Model, error) { if params.ModelType == 1706077826 { innerModel.ModelType = strconv.FormatInt(params.ModelType, 10) return innerModel, nil, nil @@ -2161,7 +2160,7 @@ func TestStreamableToolWithMultipleInterrupts(t *testing.T) { }, } - r.modelManage.EXPECT().GetModel(gomock.Any(), gomock.Any()).DoAndReturn(func(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *crossmodelmgr.Model, error) { + r.modelManage.EXPECT().GetModel(gomock.Any(), gomock.Any()).DoAndReturn(func(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *modelmgr.Model, error) { if params.ModelType == 1706077827 { outerModel.ModelType = strconv.FormatInt(params.ModelType, 10) return outerModel, nil, nil @@ -2455,7 +2454,7 @@ func TestAggregateStreamVariables(t *testing.T) { }, } - r.modelManage.EXPECT().GetModel(gomock.Any(), gomock.Any()).DoAndReturn(func(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *crossmodelmgr.Model, error) { + r.modelManage.EXPECT().GetModel(gomock.Any(), gomock.Any()).DoAndReturn(func(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *modelmgr.Model, error) { if params.ModelType == 1737521813 { cm1.ModelType = strconv.FormatInt(params.ModelType, 10) return cm1, nil, nil @@ -2598,7 +2597,7 @@ func TestParallelInterrupts(t *testing.T) { } }, } - r.modelManage.EXPECT().GetModel(gomock.Any(), gomock.Any()).DoAndReturn(func(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *crossmodelmgr.Model, error) { + r.modelManage.EXPECT().GetModel(gomock.Any(), gomock.Any()).DoAndReturn(func(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *modelmgr.Model, error) { if params.ModelType == 1737521813 { return chatModel1, nil, nil } else { @@ -3871,7 +3870,7 @@ func TestLLMException(t *testing.T) { }, } - r.modelManage.EXPECT().GetModel(gomock.Any(), gomock.Any()).DoAndReturn(func(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *crossmodelmgr.Model, error) { + r.modelManage.EXPECT().GetModel(gomock.Any(), gomock.Any()).DoAndReturn(func(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *modelmgr.Model, error) { if params.ModelType == 1737521813 { return mainChatModel, nil, nil } else { @@ -3938,7 +3937,7 @@ func TestLLMExceptionThenThrow(t *testing.T) { }, } - r.modelManage.EXPECT().GetModel(gomock.Any(), gomock.Any()).DoAndReturn(func(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *crossmodelmgr.Model, error) { + r.modelManage.EXPECT().GetModel(gomock.Any(), gomock.Any()).DoAndReturn(func(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *modelmgr.Model, error) { if params.ModelType == 1737521813 { return mainChatModel, nil, nil } else { diff --git a/backend/application/application.go b/backend/application/application.go index 5de77542..4071e486 100644 --- a/backend/application/application.go +++ b/backend/application/application.go @@ -47,7 +47,6 @@ import ( "github.com/coze-dev/coze-studio/backend/crossdomain/contract/crossdatacopy" "github.com/coze-dev/coze-studio/backend/crossdomain/contract/crossknowledge" "github.com/coze-dev/coze-studio/backend/crossdomain/contract/crossmessage" - "github.com/coze-dev/coze-studio/backend/crossdomain/contract/crossmodelmgr" "github.com/coze-dev/coze-studio/backend/crossdomain/contract/crossplugin" "github.com/coze-dev/coze-studio/backend/crossdomain/contract/crossuser" "github.com/coze-dev/coze-studio/backend/crossdomain/contract/crossvariables" @@ -60,7 +59,6 @@ import ( dataCopyImpl "github.com/coze-dev/coze-studio/backend/crossdomain/impl/datacopy" knowledgeImpl "github.com/coze-dev/coze-studio/backend/crossdomain/impl/knowledge" messageImpl "github.com/coze-dev/coze-studio/backend/crossdomain/impl/message" - modelmgrImpl "github.com/coze-dev/coze-studio/backend/crossdomain/impl/modelmgr" pluginImpl "github.com/coze-dev/coze-studio/backend/crossdomain/impl/plugin" searchImpl "github.com/coze-dev/coze-studio/backend/crossdomain/impl/search" singleagentImpl "github.com/coze-dev/coze-studio/backend/crossdomain/impl/singleagent" @@ -130,7 +128,6 @@ func Init(ctx context.Context) (err error) { crossconnector.SetDefaultSVC(connectorImpl.InitDomainService(basicServices.connectorSVC.DomainSVC)) crossdatabase.SetDefaultSVC(databaseImpl.InitDomainService(primaryServices.memorySVC.DatabaseDomainSVC)) crossknowledge.SetDefaultSVC(knowledgeImpl.InitDomainService(primaryServices.knowledgeSVC.DomainSVC)) - crossmodelmgr.SetDefaultSVC(modelmgrImpl.InitDomainService(basicServices.modelMgrSVC.DomainSVC)) crossplugin.SetDefaultSVC(pluginImpl.InitDomainService(primaryServices.pluginSVC.DomainSVC)) crossvariables.SetDefaultSVC(variablesImpl.InitDomainService(primaryServices.memorySVC.VariablesDomainSVC)) crossworkflow.SetDefaultSVC(workflowImpl.InitDomainService(primaryServices.workflowSVC.DomainSVC)) @@ -158,10 +155,7 @@ func initBasicServices(ctx context.Context, infra *appinfra.AppDependencies, e * upload.InitService(infra.TOSClient, infra.CacheCli) openAuthSVC := openauth.InitService(infra.DB, infra.IDGenSVC) promptSVC := prompt.InitService(infra.DB, infra.IDGenSVC, e.resourceEventBus) - modelMgrSVC, err := modelmgr.InitService(infra.DB, infra.IDGenSVC, infra.TOSClient) - if err != nil { - return nil, err - } + modelMgrSVC := modelmgr.InitService(infra.ModelMgr, infra.TOSClient) connectorSVC := connector.InitService(infra.TOSClient) userSVC := user.InitService(ctx, infra.DB, infra.TOSClient, infra.IDGenSVC) templateSVC := template.InitService(ctx, &template.ServiceComponents{ @@ -285,7 +279,7 @@ func (b *basicServices) toWorkflowServiceComponents(pluginSVC *plugin.PluginAppl VariablesDomainSVC: memorySVC.VariablesDomainSVC, PluginDomainSVC: pluginSVC.DomainSVC, KnowledgeDomainSVC: knowledgeSVC.DomainSVC, - ModelManager: b.modelMgrSVC.DomainSVC, + ModelManager: b.infra.ModelMgr, DomainNotifier: b.eventbus.resourceEventBus, CPStore: checkpoint.NewRedisStore(b.infra.CacheCli), } @@ -299,7 +293,7 @@ func (p *primaryServices) toSingleAgentServiceComponents() *singleagent.ServiceC Cache: p.basicServices.infra.CacheCli, TosClient: p.basicServices.infra.TOSClient, ImageX: p.basicServices.infra.ImageXClient, - ModelMgrDomainSVC: p.basicServices.modelMgrSVC.DomainSVC, + ModelMgr: p.infra.ModelMgr, UserDomainSVC: p.basicServices.userSVC.DomainSVC, EventBus: p.basicServices.eventbus.projectEventBus, DatabaseDomainSVC: p.memorySVC.DatabaseDomainSVC, diff --git a/backend/application/base/appinfra/app_infra.go b/backend/application/base/appinfra/app_infra.go index b69b2733..766d13dc 100644 --- a/backend/application/base/appinfra/app_infra.go +++ b/backend/application/base/appinfra/app_infra.go @@ -24,6 +24,7 @@ import ( "gorm.io/gorm" "github.com/coze-dev/coze-studio/backend/infra/contract/imagex" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" "github.com/coze-dev/coze-studio/backend/infra/impl/cache/redis" "github.com/coze-dev/coze-studio/backend/infra/impl/es" "github.com/coze-dev/coze-studio/backend/infra/impl/eventbus" @@ -43,6 +44,7 @@ type AppDependencies struct { TOSClient storage.Storage ResourceEventProducer eventbus.Producer AppEventProducer eventbus.Producer + ModelMgr modelmgr.Manager } func Init(ctx context.Context) (*AppDependencies, error) { @@ -86,6 +88,11 @@ func Init(ctx context.Context) (*AppDependencies, error) { return nil, err } + deps.ModelMgr, err = initModelMgr() + if err != nil { + return nil, err + } + return deps, nil } diff --git a/backend/application/modelmgr/init_by_env.go b/backend/application/base/appinfra/modelmgr.go similarity index 58% rename from backend/application/modelmgr/init_by_env.go rename to backend/application/base/appinfra/modelmgr.go index a06e006c..f32cad60 100644 --- a/backend/application/modelmgr/init_by_env.go +++ b/backend/application/base/appinfra/modelmgr.go @@ -1,21 +1,61 @@ -package modelmgr +package appinfra import ( "fmt" "os" "path/filepath" "strconv" + "strings" "gopkg.in/yaml.v3" - "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" "github.com/coze-dev/coze-studio/backend/infra/contract/chatmodel" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" + "github.com/coze-dev/coze-studio/backend/infra/impl/modelmgr/static" "github.com/coze-dev/coze-studio/backend/pkg/logs" ) -func initModelByEnv(wd, templatePath string) (metaSlice []*modelmgr.ModelMeta, entitySlice []*modelmgr.Model, err error) { - metaRoot := filepath.Join(wd, templatePath, "meta") - entityRoot := filepath.Join(wd, templatePath, "entity") +func initModelMgr() (modelmgr.Manager, error) { + wd, err := os.Getwd() + if err != nil { + return nil, err + } + + staticModel, err := initModelByTemplate(wd, "resources/conf/model") + if err != nil { + return nil, err + } + + envModel, err := initModelByEnv(wd, "resources/conf/model/template") + if err != nil { + return nil, err + } + + all := append(staticModel, envModel...) + if err := fillModelContent(all); err != nil { + return nil, err + } + + mgr, err := static.NewModelMgr(all) + if err != nil { + return nil, err + } + + return mgr, nil +} + +func initModelByTemplate(wd, configPath string) ([]*modelmgr.Model, error) { + configRoot := filepath.Join(wd, configPath) + staticModel, err := readDirYaml[modelmgr.Model](configRoot) + if err != nil { + return nil, err + } + + return staticModel, nil +} + +func initModelByEnv(wd, templatePath string) (modelEntities []*modelmgr.Model, err error) { + entityRoot := filepath.Join(wd, templatePath) for i := -1; i < 1000; i++ { rawProtocol := os.Getenv(concatEnvKey(modelProtocolPrefix, i)) @@ -35,7 +75,7 @@ func initModelByEnv(wd, templatePath string) (metaSlice []*modelmgr.ModelMeta, e mapping, found := modelMapping[protocol] if !found { - return nil, nil, fmt.Errorf("[initModelByEnv] unsupport protocol: %s", rawProtocol) + return nil, fmt.Errorf("[initModelByEnv] unsupport protocol: %s", rawProtocol) } switch protocol { @@ -44,41 +84,28 @@ func initModelByEnv(wd, templatePath string) (metaSlice []*modelmgr.ModelMeta, e if !foundTemplate { logs.Warnf("[initModelByEnv] unsupport model=%s, using default config", info.modelName) } - modelMeta, err := readYaml[modelmgr.ModelMeta](filepath.Join(metaRoot, concatTemplateFileName("model_meta_template_ark", fileSuffix))) + modelEntity, err := readYaml[modelmgr.Model](filepath.Join(entityRoot, concatTemplateFileName("model_template_ark", fileSuffix))) if err != nil { - return nil, nil, err - } - modelEntity, err := readYaml[modelmgr.Model](filepath.Join(entityRoot, concatTemplateFileName("model_entity_template_ark", fileSuffix))) - if err != nil { - return nil, nil, err + return nil, err } id, err := strconv.ParseInt(info.id, 10, 64) if err != nil { - return nil, nil, err + return nil, err } - // meta 和 entity 用一个 id，有概率冲突 - modelMeta.ID = id - modelMeta.ConnConfig.Model = info.modelID - modelMeta.ConnConfig.APIKey = info.apiKey - if info.baseURL != "" { - modelMeta.ConnConfig.BaseURL = info.baseURL - } modelEntity.ID = id - modelEntity.Meta.ID = id if !foundTemplate { modelEntity.Name = info.modelName } - metaSlice = append(metaSlice, modelMeta) - entitySlice = append(entitySlice, modelEntity) + modelEntities = append(modelEntities, modelEntity) default: - return nil, nil, fmt.Errorf("[initModelByEnv] unsupport protocol: %s", rawProtocol) + return nil, fmt.Errorf("[initModelByEnv] unsupport protocol: %s", rawProtocol) } } - return metaSlice, entitySlice, nil + return modelEntities, nil } type envModelInfo struct { @@ -95,6 +122,32 @@ func getModelEnv(idx int) (info envModelInfo, valid bool) { return } +func readDirYaml[T any](dir string) ([]*T, error) { + des, err := os.ReadDir(dir) + if err != nil { + return nil, err + } + resp := make([]*T, 0, len(des)) + for _, file := range des { + if file.IsDir() { + continue + } + if strings.HasSuffix(file.Name(), ".yaml") || strings.HasSuffix(file.Name(), ".yml") { + filePath := filepath.Join(dir, file.Name()) + data, err := os.ReadFile(filePath) + if err != nil { + return nil, err + } + var content T + if err := yaml.Unmarshal(data, &content); err != nil { + return nil, err + } + resp = append(resp, &content) + } + } + return resp, nil +} + func readYaml[T any](fPath string) (*T, error) { data, err := os.ReadFile(fPath) if err != nil { @@ -146,3 +199,18 @@ var modelMapping = map[chatmodel.Protocol]map[string]string{ "deepseek-v3": "volc_deepseek-v3", }, } + +func fillModelContent(items []*modelmgr.Model) error { + for i := range items { + item := items[i] + if item.Meta.Status == modelmgr.StatusDefault { + item.Meta.Status = modelmgr.StatusInUse + } + + if item.IconURI == "" && item.IconURL == "" { + return fmt.Errorf("missing icon URI or icon URL, id=%d", item.ID) + } + } + + return nil +} diff --git a/backend/application/modelmgr/init_by_env_test.go b/backend/application/base/appinfra/modelmgr_test.go similarity index 83% rename from backend/application/modelmgr/init_by_env_test.go rename to backend/application/base/appinfra/modelmgr_test.go index cadbdfb5..6fe4bd1f 100644 --- a/backend/application/modelmgr/init_by_env_test.go +++ b/backend/application/base/appinfra/modelmgr_test.go @@ -1,4 +1,4 @@ -package modelmgr +package appinfra import ( "fmt" @@ -24,8 +24,7 @@ func TestInitByEnv(t *testing.T) { wd, err := os.Getwd() assert.NoError(t, err) - ms, es, err := initModelByEnv(wd, "../../conf/model/template") + ms, err := initModelByEnv(wd, "../../../conf/model/template") assert.NoError(t, err) assert.Len(t, ms, len(modelMapping[chatmodel.ProtocolArk])) - assert.Len(t, es, len(modelMapping[chatmodel.ProtocolArk])) } diff --git a/backend/application/modelmgr/init.go b/backend/application/modelmgr/init.go index 7ded533c..40c8bfc9 100644 --- a/backend/application/modelmgr/init.go +++ b/backend/application/modelmgr/init.go @@ -1,198 +1,11 @@ -/* - * Copyright 2025 coze-dev Authors - * - * Licensed under the Apache License, Version 2.0 (the "License"); - * you may not use this file except in compliance with the License. - * You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - package modelmgr import ( - "context" - "errors" - "fmt" - "os" - "path/filepath" - "strings" - "time" - - "gopkg.in/yaml.v3" - "gorm.io/gorm" - - crossmodelmgr "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr/entity" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr/service" - "github.com/coze-dev/coze-studio/backend/infra/contract/storage" - "github.com/coze-dev/coze-studio/backend/infra/impl/idgen" - "github.com/coze-dev/coze-studio/backend/pkg/logs" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" + "github.com/coze-dev/coze-studio/backend/infra/impl/storage" ) -func InitService(db *gorm.DB, idgen idgen.IDGenerator, oss storage.Storage) (*ModelmgrApplicationService, error) { - svc := service.NewModelManager(db, idgen, oss) - if err := loadStaticModelConfig(svc, oss); err != nil { - return nil, err - } - ModelmgrApplicationSVC.DomainSVC = svc - - return ModelmgrApplicationSVC, nil -} - -func loadStaticModelConfig(svc modelmgr.Manager, oss storage.Storage) error { - ctx := context.Background() - - id2Meta := make(map[int64]*entity.ModelMeta) - var cursor *string - for { - req := &modelmgr.ListModelMetaRequest{ - Status: []entity.ModelMetaStatus{ - crossmodelmgr.StatusInUse, - crossmodelmgr.StatusPending, - crossmodelmgr.StatusDeleted, - }, - Limit: 100, - Cursor: cursor, - } - listMetaResp, err := svc.ListModelMeta(ctx, req) - if err != nil { - return err - } - for _, item := range listMetaResp.ModelMetaList { - cpItem := item - id2Meta[cpItem.ID] = cpItem - } - if !listMetaResp.HasMore { - break - } - cursor = listMetaResp.NextCursor - } - - root, err := os.Getwd() - if err != nil { - return err - } - - envModelMeta, envModelEntity, err := initModelByEnv(root, "resources/conf/model/template") - if err != nil { - return err - } - - filePath := filepath.Join(root, "resources/conf/model/meta") - staticModelMeta, err := readDirYaml[crossmodelmgr.ModelMeta](filePath) - if err != nil { - return err - } - staticModelMeta = append(staticModelMeta, envModelMeta...) - for _, modelMeta := range staticModelMeta { - if _, found := id2Meta[modelMeta.ID]; !found { - if modelMeta.IconURI == "" && modelMeta.IconURL == "" { - return fmt.Errorf("missing icon URI or icon URL, id=%d", modelMeta.ID) - } else if modelMeta.IconURL != "" { - // do nothing - } else if modelMeta.IconURI != "" { - // try local path - base := filepath.Base(modelMeta.IconURI) - iconPath := filepath.Join("resources/conf/model/icon", base) - if _, err = os.Stat(iconPath); err == nil { - // try upload icon - icon, err := os.ReadFile(iconPath) - if err != nil { - return err - } - key := fmt.Sprintf("icon_%s_%d", base, time.Now().Second()) - if err := oss.PutObject(ctx, key, icon); err != nil { - return err - } - modelMeta.IconURI = key - } else if errors.Is(err, os.ErrNotExist) { - // try to get object from uri - if _, err := oss.GetObject(ctx, modelMeta.IconURI); err != nil { - return err - } - } else { - return err - } - } - newMeta, err := svc.CreateModelMeta(ctx, modelMeta) - if err != nil { - if errors.Is(err, gorm.ErrDuplicatedKey) { - logs.Infof("[loadStaticModelConfig] model meta conflict for id=%d, skip", newMeta.ID) - } - return err - } else { - logs.Infof("[loadStaticModelConfig] model meta create success, id=%d", newMeta.ID) - } - id2Meta[newMeta.ID] = newMeta - } else { - logs.Infof("[loadStaticModelConfig] model meta founded, skip create, id=%d", modelMeta.ID) - - } - } - - filePath = filepath.Join(root, "resources/conf/model/entity") - staticModel, err := readDirYaml[crossmodelmgr.Model](filePath) - if err != nil { - return err - } - staticModel = append(staticModel, envModelEntity...) - for _, modelEntity := range staticModel { - curModelEntities, err := svc.MGetModelByID(ctx, &modelmgr.MGetModelRequest{IDs: []int64{modelEntity.ID}}) - if err != nil { - return err - } - if len(curModelEntities) > 0 { - logs.Infof("[loadStaticModelConfig] model entity founded, skip create, id=%d", modelEntity.ID) - continue - } - meta, found := id2Meta[modelEntity.Meta.ID] - if !found { - return fmt.Errorf("model meta not found for id=%d, model_id=%d", modelEntity.Meta.ID, modelEntity.ID) - } - modelEntity.Meta = *meta - if _, err = svc.CreateModel(ctx, &entity.Model{Model: modelEntity}); err != nil { - if errors.Is(err, gorm.ErrDuplicatedKey) { - logs.Infof("[loadStaticModelConfig] model entity conflict for id=%d, skip", modelEntity.ID) - } - return err - } else { - logs.Infof("[loadStaticModelConfig] model entity create success, id=%d", modelEntity.ID) - } - } - - return nil -} - -func readDirYaml[T any](dir string) ([]*T, error) { - des, err := os.ReadDir(dir) - if err != nil { - return nil, err - } - resp := make([]*T, 0, len(des)) - for _, file := range des { - if file.IsDir() { - continue - } - if strings.HasSuffix(file.Name(), ".yaml") || strings.HasSuffix(file.Name(), ".yml") { - filePath := filepath.Join(dir, file.Name()) - data, err := os.ReadFile(filePath) - if err != nil { - return nil, err - } - var content T - if err := yaml.Unmarshal(data, &content); err != nil { - return nil, err - } - resp = append(resp, &content) - } - } - return resp, nil +func InitService(mgr modelmgr.Manager, tosClient storage.Storage) *ModelmgrApplicationService { + ModelmgrApplicationSVC = &ModelmgrApplicationService{mgr, tosClient} + return ModelmgrApplicationSVC } diff --git a/backend/application/modelmgr/modelmgr.go b/backend/application/modelmgr/modelmgr.go index 31318f5b..5bd575f3 100644 --- a/backend/application/modelmgr/modelmgr.go +++ b/backend/application/modelmgr/modelmgr.go @@ -19,10 +19,9 @@ package modelmgr import ( "context" - modelmgrEntity "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" "github.com/coze-dev/coze-studio/backend/api/model/ocean/cloud/developer_api" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr" - modelEntity "github.com/coze-dev/coze-studio/backend/domain/modelmgr/entity" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" + "github.com/coze-dev/coze-studio/backend/infra/impl/storage" "github.com/coze-dev/coze-studio/backend/pkg/i18n" "github.com/coze-dev/coze-studio/backend/pkg/lang/ptr" "github.com/coze-dev/coze-studio/backend/pkg/lang/sets" @@ -31,18 +30,19 @@ import ( ) type ModelmgrApplicationService struct { - DomainSVC modelmgr.Manager + Mgr modelmgr.Manager + TosClient storage.Storage } var ModelmgrApplicationSVC = &ModelmgrApplicationService{} -func (m *ModelmgrApplicationService) GetModelList(ctx context.Context, req *developer_api.GetTypeListRequest) ( +func (m *ModelmgrApplicationService) GetModelList(ctx context.Context, _ *developer_api.GetTypeListRequest) ( resp *developer_api.GetTypeListResponse, err error, ) { // 一般不太可能同时配置这么多模型 const modelMaxLimit = 300 - modelResp, err := m.DomainSVC.ListModel(ctx, &modelmgr.ListModelRequest{ + modelResp, err := m.Mgr.ListModel(ctx, &modelmgr.ListModelRequest{ Limit: modelMaxLimit, Cursor: nil, }) @@ -51,9 +51,15 @@ func (m *ModelmgrApplicationService) GetModelList(ctx context.Context, req *deve } locale := i18n.GetLocale(ctx) - modelList, err := slices.TransformWithErrorCheck(modelResp.ModelList, func(m *modelEntity.Model) (*developer_api.Model, error) { - logs.CtxInfof(ctx, "ChatModel DefaultParameters: %v", m.DefaultParameters) - return modelDo2To(m, locale) + modelList, err := slices.TransformWithErrorCheck(modelResp.ModelList, func(mm *modelmgr.Model) (*developer_api.Model, error) { + logs.CtxInfof(ctx, "ChatModel DefaultParameters: %v", mm.DefaultParameters) + if mm.IconURI != "" { + iconUrl, err := m.TosClient.GetObjectUrl(ctx, mm.IconURI) + if err == nil { + mm.IconURL = iconUrl + } + } + return modelDo2To(mm, locale) }) if err != nil { return nil, err @@ -68,11 +74,11 @@ func (m *ModelmgrApplicationService) GetModelList(ctx context.Context, req *deve }, nil } -func modelDo2To(model *modelEntity.Model, locale i18n.Locale) (*developer_api.Model, error) { +func modelDo2To(model *modelmgr.Model, locale i18n.Locale) (*developer_api.Model, error) { mm := model.Meta mps := slices.Transform(model.DefaultParameters, - func(param *modelmgrEntity.Parameter) *developer_api.ModelParameter { + func(param *modelmgr.Parameter) *developer_api.ModelParameter { return parameterDo2To(param, locale) }, ) @@ -83,7 +89,7 @@ func modelDo2To(model *modelEntity.Model, locale i18n.Locale) (*developer_api.Mo Name: model.Name, ModelType: model.ID, ModelClass: mm.Protocol.TOModelClass(), - ModelIcon: mm.IconURL, + ModelIcon: model.IconURL, ModelInputPrice: 0, ModelOutputPrice: 0, ModelQuota: &developer_api.ModelQuota{ @@ -102,19 +108,19 @@ func modelDo2To(model *modelEntity.Model, locale i18n.Locale) (*developer_api.Mo }, ModelName: mm.Name, ModelClassName: mm.Protocol.TOModelClass().String(), - IsOffline: mm.Status != modelmgrEntity.StatusInUse, + IsOffline: mm.Status != modelmgr.StatusInUse, ModelParams: mps, ModelDesc: []*developer_api.ModelDescGroup{ { GroupName: "Description", - Desc: []string{model.Description}, + Desc: []string{model.Description.Read(locale)}, }, }, FuncConfig: nil, EndpointName: nil, ModelTagList: nil, IsUpRequired: nil, - ModelBriefDesc: mm.Description.Read(locale), + ModelBriefDesc: model.Description.Read(locale), ModelSeries: &developer_api.ModelSeriesInfo{ // TODO: 替换为真实配置 SeriesName: "热门模型", }, @@ -122,16 +128,16 @@ func modelDo2To(model *modelEntity.Model, locale i18n.Locale) (*developer_api.Mo ModelAbility: &developer_api.ModelAbility{ CotDisplay: ptr.Of(mm.Capability.Reasoning), FunctionCall: ptr.Of(mm.Capability.FunctionCall), - ImageUnderstanding: ptr.Of(modalSet.Contains(modelmgrEntity.ModalImage)), - VideoUnderstanding: ptr.Of(modalSet.Contains(modelmgrEntity.ModalVideo)), - AudioUnderstanding: ptr.Of(modalSet.Contains(modelmgrEntity.ModalAudio)), + ImageUnderstanding: ptr.Of(modalSet.Contains(modelmgr.ModalImage)), + VideoUnderstanding: ptr.Of(modalSet.Contains(modelmgr.ModalVideo)), + AudioUnderstanding: ptr.Of(modalSet.Contains(modelmgr.ModalAudio)), SupportMultiModal: ptr.Of(len(modalSet) > 1), PrefillResp: ptr.Of(mm.Capability.PrefillResponse), }, }, nil } -func parameterDo2To(param *modelmgrEntity.Parameter, locale i18n.Locale) *developer_api.ModelParameter { +func parameterDo2To(param *modelmgr.Parameter, locale i18n.Locale) *developer_api.ModelParameter { if param == nil { return nil } @@ -146,19 +152,19 @@ func parameterDo2To(param *modelmgrEntity.Parameter, locale i18n.Locale) *develo var custom string var creative, balance, precise *string - if val, ok := param.DefaultVal[modelmgrEntity.DefaultTypeDefault]; ok { + if val, ok := param.DefaultVal[modelmgr.DefaultTypeDefault]; ok { custom = val } - if val, ok := param.DefaultVal[modelmgrEntity.DefaultTypeCreative]; ok { + if val, ok := param.DefaultVal[modelmgr.DefaultTypeCreative]; ok { creative = ptr.Of(val) } - if val, ok := param.DefaultVal[modelmgrEntity.DefaultTypeBalance]; ok { + if val, ok := param.DefaultVal[modelmgr.DefaultTypeBalance]; ok { balance = ptr.Of(val) } - if val, ok := param.DefaultVal[modelmgrEntity.DefaultTypePrecise]; ok { + if val, ok := param.DefaultVal[modelmgr.DefaultTypePrecise]; ok { precise = ptr.Of(val) } @@ -168,11 +174,11 @@ func parameterDo2To(param *modelmgrEntity.Parameter, locale i18n.Locale) *develo Desc: param.Desc.Read(locale), Type: func() developer_api.ModelParamType { switch param.Type { - case modelmgrEntity.ValueTypeBoolean: + case modelmgr.ValueTypeBoolean: return developer_api.ModelParamType_Boolean - case modelmgrEntity.ValueTypeInt: + case modelmgr.ValueTypeInt: return developer_api.ModelParamType_Int - case modelmgrEntity.ValueTypeFloat: + case modelmgr.ValueTypeFloat: return developer_api.ModelParamType_Float default: return developer_api.ModelParamType_String @@ -191,9 +197,9 @@ func parameterDo2To(param *modelmgrEntity.Parameter, locale i18n.Locale) *develo ParamClass: &developer_api.ModelParamClass{ ClassID: func() int32 { switch param.Style.Widget { - case modelmgrEntity.WidgetSlider: + case modelmgr.WidgetSlider: return 1 - case modelmgrEntity.WidgetRadioButtons: + case modelmgr.WidgetRadioButtons: return 2 default: return 0 diff --git a/backend/application/singleagent/create.go b/backend/application/singleagent/create.go index 10a70562..677ee6a0 100644 --- a/backend/application/singleagent/create.go +++ b/backend/application/singleagent/create.go @@ -20,15 +20,14 @@ import ( "context" "time" - modelmgrEntity "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/singleagent" intelligence "github.com/coze-dev/coze-studio/backend/api/model/intelligence/common" "github.com/coze-dev/coze-studio/backend/api/model/ocean/cloud/bot_common" "github.com/coze-dev/coze-studio/backend/api/model/ocean/cloud/developer_api" "github.com/coze-dev/coze-studio/backend/application/base/ctxutil" "github.com/coze-dev/coze-studio/backend/domain/agent/singleagent/entity" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr" searchEntity "github.com/coze-dev/coze-studio/backend/domain/search/entity" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" "github.com/coze-dev/coze-studio/backend/pkg/errorx" "github.com/coze-dev/coze-studio/backend/pkg/lang/ptr" "github.com/coze-dev/coze-studio/backend/types/errno" @@ -95,7 +94,7 @@ func (s *SingleAgentApplicationService) newDefaultSingleAgent(ctx context.Contex Plugin: []*bot_common.PluginInfo{}, Knowledge: &bot_common.Knowledge{ TopK: ptr.Of(int64(1)), - MinScore: ptr.Of(float64(0.01)), + MinScore: ptr.Of(0.01), SearchStrategy: ptr.Of(bot_common.SearchStrategy_SemanticSearch), RecallStrategy: &bot_common.RecallStrategy{ UseNl2sql: ptr.Of(true), @@ -115,8 +114,8 @@ func (s *SingleAgentApplicationService) newDefaultSingleAgent(ctx context.Contex } func (s *SingleAgentApplicationService) defaultModelInfo(ctx context.Context) (*bot_common.ModelInfo, error) { - modelResp, err := s.appContext.ModelMgrDomainSVC.ListModel(ctx, &modelmgr.ListModelRequest{ - Status: []modelmgrEntity.ModelEntityStatus{modelmgrEntity.ModelEntityStatusDefault, modelmgrEntity.ModelEntityStatusInUse}, + modelResp, err := s.appContext.ModelMgr.ListModel(ctx, &modelmgr.ListModelRequest{ + Status: []modelmgr.ModelStatus{modelmgr.StatusInUse}, Limit: 1, Cursor: nil, }) @@ -131,8 +130,8 @@ func (s *SingleAgentApplicationService) defaultModelInfo(ctx context.Context) (* dm := modelResp.ModelList[0] var temperature *float64 - if tp, ok := dm.FindParameter(modelmgrEntity.Temperature); ok { - t, err := tp.GetFloat(modelmgrEntity.DefaultTypeBalance) + if tp, ok := dm.FindParameter(modelmgr.Temperature); ok { + t, err := tp.GetFloat(modelmgr.DefaultTypeBalance) if err != nil { return nil, err } @@ -141,8 +140,8 @@ func (s *SingleAgentApplicationService) defaultModelInfo(ctx context.Context) (* } var maxTokens *int32 - if tp, ok := dm.FindParameter(modelmgrEntity.MaxTokens); ok { - t, err := tp.GetInt(modelmgrEntity.DefaultTypeBalance) + if tp, ok := dm.FindParameter(modelmgr.MaxTokens); ok { + t, err := tp.GetInt(modelmgr.DefaultTypeBalance) if err != nil { return nil, err } @@ -152,8 +151,8 @@ func (s *SingleAgentApplicationService) defaultModelInfo(ctx context.Context) (* } var topP *float64 - if tp, ok := dm.FindParameter(modelmgrEntity.TopP); ok { - t, err := tp.GetFloat(modelmgrEntity.DefaultTypeBalance) + if tp, ok := dm.FindParameter(modelmgr.TopP); ok { + t, err := tp.GetFloat(modelmgr.DefaultTypeBalance) if err != nil { return nil, err } @@ -161,8 +160,8 @@ func (s *SingleAgentApplicationService) defaultModelInfo(ctx context.Context) (* } var topK *int32 - if tp, ok := dm.FindParameter(modelmgrEntity.TopK); ok { - t, err := tp.GetInt(modelmgrEntity.DefaultTypeBalance) + if tp, ok := dm.FindParameter(modelmgr.TopK); ok { + t, err := tp.GetInt(modelmgr.DefaultTypeBalance) if err != nil { return nil, err } @@ -170,8 +169,8 @@ func (s *SingleAgentApplicationService) defaultModelInfo(ctx context.Context) (* } var frequencyPenalty *float64 - if tp, ok := dm.FindParameter(modelmgrEntity.FrequencyPenalty); ok { - t, err := tp.GetFloat(modelmgrEntity.DefaultTypeBalance) + if tp, ok := dm.FindParameter(modelmgr.FrequencyPenalty); ok { + t, err := tp.GetFloat(modelmgr.DefaultTypeBalance) if err != nil { return nil, err } @@ -179,8 +178,8 @@ func (s *SingleAgentApplicationService) defaultModelInfo(ctx context.Context) (* } var presencePenalty *float64 - if tp, ok := dm.FindParameter(modelmgrEntity.PresencePenalty); ok { - t, err := tp.GetFloat(modelmgrEntity.DefaultTypeBalance) + if tp, ok := dm.FindParameter(modelmgr.PresencePenalty); ok { + t, err := tp.GetFloat(modelmgr.DefaultTypeBalance) if err != nil { return nil, err } diff --git a/backend/application/singleagent/get.go b/backend/application/singleagent/get.go index 289fe351..3cc8bfe7 100644 --- a/backend/application/singleagent/get.go +++ b/backend/application/singleagent/get.go @@ -30,13 +30,12 @@ import ( "github.com/coze-dev/coze-studio/backend/api/model/plugin_develop_common" "github.com/coze-dev/coze-studio/backend/domain/agent/singleagent/entity" knowledge "github.com/coze-dev/coze-studio/backend/domain/knowledge/service" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr" - modelEntity "github.com/coze-dev/coze-studio/backend/domain/modelmgr/entity" pluginEntity "github.com/coze-dev/coze-studio/backend/domain/plugin/entity" "github.com/coze-dev/coze-studio/backend/domain/plugin/service" shortcutCMDEntity "github.com/coze-dev/coze-studio/backend/domain/shortcutcmd/entity" workflowEntity "github.com/coze-dev/coze-studio/backend/domain/workflow/entity" "github.com/coze-dev/coze-studio/backend/domain/workflow/entity/vo" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" "github.com/coze-dev/coze-studio/backend/pkg/errorx" "github.com/coze-dev/coze-studio/backend/pkg/lang/conv" "github.com/coze-dev/coze-studio/backend/pkg/lang/ptr" @@ -159,13 +158,13 @@ func (s *SingleAgentApplicationService) shortcutCMDDo2Vo(cmdDOs []*shortcutCMDEn }) } -func (s *SingleAgentApplicationService) fetchModelDetails(ctx context.Context, agentInfo *entity.SingleAgent) ([]*modelEntity.Model, error) { +func (s *SingleAgentApplicationService) fetchModelDetails(ctx context.Context, agentInfo *entity.SingleAgent) ([]*modelmgr.Model, error) { if agentInfo.ModelInfo.ModelId == nil { return nil, nil } modelID := agentInfo.ModelInfo.GetModelId() - modelInfos, err := s.appContext.ModelMgrDomainSVC.MGetModelByID(ctx, &modelmgr.MGetModelRequest{ + modelInfos, err := s.appContext.ModelMgr.MGetModelByID(ctx, &modelmgr.MGetModelRequest{ IDs: []int64{modelID}, }) if err != nil { @@ -249,13 +248,13 @@ func (s *SingleAgentApplicationService) fetchWorkflowDetails(ctx context.Context return ret, nil } -func modelInfoDo2Vo(modelInfos []*modelEntity.Model) map[int64]*playground.ModelDetail { - return slices.ToMap(modelInfos, func(e *modelEntity.Model) (int64, *playground.ModelDetail) { +func modelInfoDo2Vo(modelInfos []*modelmgr.Model) map[int64]*playground.ModelDetail { + return slices.ToMap(modelInfos, func(e *modelmgr.Model) (int64, *playground.ModelDetail) { return e.ID, toModelDetail(e) }) } -func toModelDetail(m *modelEntity.Model) *playground.ModelDetail { +func toModelDetail(m *modelmgr.Model) *playground.ModelDetail { mm := m.Meta return &playground.ModelDetail{ @@ -263,7 +262,7 @@ func toModelDetail(m *modelEntity.Model) *playground.ModelDetail { ModelName: ptr.Of(m.Meta.Name), ModelID: ptr.Of(m.ID), ModelFamily: ptr.Of(int64(mm.Protocol.TOModelClass())), - ModelIconURL: ptr.Of(mm.IconURL), + ModelIconURL: ptr.Of(m.IconURL), } } diff --git a/backend/application/singleagent/init.go b/backend/application/singleagent/init.go index c8c4bb4e..660c2b01 100644 --- a/backend/application/singleagent/init.go +++ b/backend/application/singleagent/init.go @@ -28,7 +28,6 @@ import ( knowledge "github.com/coze-dev/coze-studio/backend/domain/knowledge/service" database "github.com/coze-dev/coze-studio/backend/domain/memory/database/service" variables "github.com/coze-dev/coze-studio/backend/domain/memory/variables/service" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr" "github.com/coze-dev/coze-studio/backend/domain/plugin/service" search "github.com/coze-dev/coze-studio/backend/domain/search/service" shortcutCmd "github.com/coze-dev/coze-studio/backend/domain/shortcutcmd/service" @@ -36,6 +35,7 @@ import ( "github.com/coze-dev/coze-studio/backend/domain/workflow" "github.com/coze-dev/coze-studio/backend/infra/contract/idgen" "github.com/coze-dev/coze-studio/backend/infra/contract/imagex" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" "github.com/coze-dev/coze-studio/backend/infra/contract/storage" "github.com/coze-dev/coze-studio/backend/infra/impl/chatmodel" "github.com/coze-dev/coze-studio/backend/pkg/jsoncache" @@ -55,9 +55,9 @@ type ServiceComponents struct { ImageX imagex.ImageX EventBus search.ProjectEventBus CounterRepo repository.CounterRepository + ModelMgr modelmgr.Manager KnowledgeDomainSVC knowledge.Knowledge - ModelMgrDomainSVC modelmgr.Manager PluginDomainSVC service.PluginService WorkflowDomainSVC workflow.Service UserDomainSVC user.User @@ -76,6 +76,7 @@ func InitService(c *ServiceComponents) (*SingleAgentApplicationService, error) { CounterRepo: repository.NewCounterRepo(c.Cache), CPStore: c.CPStore, ModelFactory: chatmodel.NewDefaultFactory(), + ModelMgr: c.ModelMgr, } singleAgentDomainSVC := singleagent.NewService(domainComponents) diff --git a/backend/application/workflow/init.go b/backend/application/workflow/init.go index 9d7c7b29..6a6d0a76 100644 --- a/backend/application/workflow/init.go +++ b/backend/application/workflow/init.go @@ -30,7 +30,6 @@ import ( knowledge "github.com/coze-dev/coze-studio/backend/domain/knowledge/service" dbservice "github.com/coze-dev/coze-studio/backend/domain/memory/database/service" variables "github.com/coze-dev/coze-studio/backend/domain/memory/variables/service" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr" plugin "github.com/coze-dev/coze-studio/backend/domain/plugin/service" search "github.com/coze-dev/coze-studio/backend/domain/search/service" "github.com/coze-dev/coze-studio/backend/domain/workflow" @@ -44,6 +43,7 @@ import ( "github.com/coze-dev/coze-studio/backend/domain/workflow/service" "github.com/coze-dev/coze-studio/backend/infra/contract/idgen" "github.com/coze-dev/coze-studio/backend/infra/contract/imagex" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" "github.com/coze-dev/coze-studio/backend/infra/contract/storage" "github.com/coze-dev/coze-studio/backend/infra/impl/coderunner" ) diff --git a/backend/conf/model/entity/.gitkeep b/backend/conf/model/entity/.gitkeep deleted file mode 100644 index e69de29b..00000000 diff --git a/backend/conf/model/meta/.gitkeep b/backend/conf/model/meta/.gitkeep deleted file mode 100644 index e69de29b..00000000 diff --git a/backend/conf/model/template/entity/model_entity_template_ark.yaml b/backend/conf/model/template/entity/model_entity_template_ark.yaml deleted file mode 100644 index c803beea..00000000 --- a/backend/conf/model/template/entity/model_entity_template_ark.yaml +++ /dev/null @@ -1,94 +0,0 @@ -id: 2002 # 模型 entity id, 同 id 数据不会覆盖 -name: Doubao Model -description: test doubao description -meta: - id: 102 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - label_en: '' - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: response_format - label: - zh: 输出格式 - en: Response format - desc: - zh: '- **文本**: 使用普通文本格式回复\n- **Markdown**: 将引导模型使用Markdown格式输出回复\n- **JSON**: 将引导模型使用JSON格式输出' - en: '**Response Format**:\n\n- **Text**: Replies in plain text format\n- **Markdown**: Uses Markdown format for replies\n- **JSON**: Uses JSON format for replies' - type: int - precision: 0 - default_val: - default_val: '0' - options: - - - value: '0' - label: 'Text' - - - value: '1' - label: 'Markdown' - - - value: '2' - label: 'JSON' - style: - widget: radio_buttons - label: - zh: 输入及输出设置 - en: Input and output settings \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-lite.yaml b/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-lite.yaml deleted file mode 100644 index 08fdc398..00000000 --- a/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-lite.yaml +++ /dev/null @@ -1,66 +0,0 @@ -id: 65536 # 模型 entity id, 同 id 数据不会覆盖 -name: Doubao-1.5-Lite -meta: - id: 65536 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - label_en: '' - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-pro-256k.yaml b/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-pro-256k.yaml deleted file mode 100644 index cb830d68..00000000 --- a/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-pro-256k.yaml +++ /dev/null @@ -1,66 +0,0 @@ -id: 65536 # 模型 entity id, 同 id 数据不会覆盖 -name: Doubao-1.5-Pro-256k -meta: - id: 65536 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - label_en: '' - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-pro-32k.yaml b/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-pro-32k.yaml deleted file mode 100644 index 4bbf9aee..00000000 --- a/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-pro-32k.yaml +++ /dev/null @@ -1,66 +0,0 @@ -id: 65536 # 模型 entity id, 同 id 数据不会覆盖 -name: Doubao-1.5-Pro-32k -meta: - id: 65536 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - label_en: '' - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-thinking-pro.yaml b/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-thinking-pro.yaml deleted file mode 100644 index bc652b4f..00000000 --- a/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-thinking-pro.yaml +++ /dev/null @@ -1,90 +0,0 @@ -id: 65536 # 模型 entity id, 同 id 数据不会覆盖 -name: Doubao-1.5-Thinking-Pro -meta: - id: 65536 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - label_en: '' - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: response_format - label: - zh: 输出格式 - en: Response format - desc: - zh: '- **JSON**: 将引导模型使用JSON格式输出' - en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' - type: int - precision: 0 - default_val: - default_val: '0' - options: - - - value: '0' - label: 'Text' - - - value: '1' - label: 'JSON' - style: - widget: radio_buttons - label: - zh: 输入及输出设置 - en: Input and output settings \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-thinking-vision-pro.yaml b/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-thinking-vision-pro.yaml deleted file mode 100644 index 73a62cc1..00000000 --- a/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-thinking-vision-pro.yaml +++ /dev/null @@ -1,90 +0,0 @@ -id: 65536 # 模型 entity id, 同 id 数据不会覆盖 -name: Doubao-1.5-Thinking-Vision-Pro -meta: - id: 65536 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - label_en: '' - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: response_format - label: - zh: 输出格式 - en: Response format - desc: - zh: '- **JSON**: 将引导模型使用JSON格式输出' - en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' - type: int - precision: 0 - default_val: - default_val: '0' - options: - - - value: '0' - label: 'Text' - - - value: '1' - label: 'JSON' - style: - widget: radio_buttons - label: - zh: 输入及输出设置 - en: Input and output settings \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-vision-lite.yaml b/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-vision-lite.yaml deleted file mode 100644 index 40ff83a5..00000000 --- a/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-vision-lite.yaml +++ /dev/null @@ -1,90 +0,0 @@ -id: 65536 # 模型 entity id, 同 id 数据不会覆盖 -name: Doubao-1.5-Vision-Lite -meta: - id: 65536 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - label_en: '' - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: response_format - label: - zh: 输出格式 - en: Response format - desc: - zh: '- **JSON**: 将引导模型使用JSON格式输出' - en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' - type: int - precision: 0 - default_val: - default_val: '0' - options: - - - value: '0' - label: 'Text' - - - value: '1' - label: 'JSON' - style: - widget: radio_buttons - label: - zh: 输入及输出设置 - en: Input and output settings \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-vision-pro.yaml b/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-vision-pro.yaml deleted file mode 100644 index 6b848726..00000000 --- a/backend/conf/model/template/entity/model_entity_template_ark_doubao-1.5-vision-pro.yaml +++ /dev/null @@ -1,90 +0,0 @@ -id: 65536 # 模型 entity id, 同 id 数据不会覆盖 -name: Doubao-1.5-Vision-Pro -meta: - id: 65536 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - label_en: '' - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: response_format - label: - zh: 输出格式 - en: Response format - desc: - zh: '- **JSON**: 将引导模型使用JSON格式输出' - en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' - type: int - precision: 0 - default_val: - default_val: '0' - options: - - - value: '0' - label: 'Text' - - - value: '1' - label: 'JSON' - style: - widget: radio_buttons - label: - zh: 输入及输出设置 - en: Input and output settings \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_ark_doubao-seed-1.6-flash.yaml b/backend/conf/model/template/entity/model_entity_template_ark_doubao-seed-1.6-flash.yaml deleted file mode 100644 index e7b0b522..00000000 --- a/backend/conf/model/template/entity/model_entity_template_ark_doubao-seed-1.6-flash.yaml +++ /dev/null @@ -1,90 +0,0 @@ -id: 65536 # 模型 entity id, 同 id 数据不会覆盖 -name: Doubao-Seed-1.6-Flash -meta: - id: 65536 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - label_en: '' - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: response_format - label: - zh: 输出格式 - en: Response format - desc: - zh: '- **JSON**: 将引导模型使用JSON格式输出' - en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' - type: int - precision: 0 - default_val: - default_val: '0' - options: - - - value: '0' - label: 'Text' - - - value: '1' - label: 'JSON' - style: - widget: radio_buttons - label: - zh: 输入及输出设置 - en: Input and output settings \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_ark_doubao-seed-1.6-thinking.yaml b/backend/conf/model/template/entity/model_entity_template_ark_doubao-seed-1.6-thinking.yaml deleted file mode 100644 index 310f692b..00000000 --- a/backend/conf/model/template/entity/model_entity_template_ark_doubao-seed-1.6-thinking.yaml +++ /dev/null @@ -1,90 +0,0 @@ -id: 65536 # 模型 entity id, 同 id 数据不会覆盖 -name: Doubao-Seed-1.6-Thinking -meta: - id: 65536 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - label_en: '' - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: response_format - label: - zh: 输出格式 - en: Response format - desc: - zh: '- **JSON**: 将引导模型使用JSON格式输出' - en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' - type: int - precision: 0 - default_val: - default_val: '0' - options: - - - value: '0' - label: 'Text' - - - value: '1' - label: 'JSON' - style: - widget: radio_buttons - label: - zh: 输入及输出设置 - en: Input and output settings \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_ark_doubao-seed-1.6.yaml b/backend/conf/model/template/entity/model_entity_template_ark_doubao-seed-1.6.yaml deleted file mode 100644 index 0a45679f..00000000 --- a/backend/conf/model/template/entity/model_entity_template_ark_doubao-seed-1.6.yaml +++ /dev/null @@ -1,90 +0,0 @@ -id: 65536 # 模型 entity id, 同 id 数据不会覆盖 -name: Doubao-Seed-1.6 -meta: - id: 65536 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - label_en: '' - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: response_format - label: - zh: 输出格式 - en: Response format - desc: - zh: '- **JSON**: 将引导模型使用JSON格式输出' - en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' - type: int - precision: 0 - default_val: - default_val: '0' - options: - - - value: '0' - label: 'Text' - - - value: '1' - label: 'JSON' - style: - widget: radio_buttons - label: - zh: 输入及输出设置 - en: Input and output settings \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_ark_volc_deepseek-r1.yaml b/backend/conf/model/template/entity/model_entity_template_ark_volc_deepseek-r1.yaml deleted file mode 100644 index 27e31eef..00000000 --- a/backend/conf/model/template/entity/model_entity_template_ark_volc_deepseek-r1.yaml +++ /dev/null @@ -1,66 +0,0 @@ -id: 65536 # 模型 entity id, 同 id 数据不会覆盖 -name: Deepseek-R1-VolcEngine -meta: - id: 65536 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - label_en: '' - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_ark_volc_deepseek-v3.yaml b/backend/conf/model/template/entity/model_entity_template_ark_volc_deepseek-v3.yaml deleted file mode 100644 index 4a47d20c..00000000 --- a/backend/conf/model/template/entity/model_entity_template_ark_volc_deepseek-v3.yaml +++ /dev/null @@ -1,66 +0,0 @@ -id: 65536 # 模型 entity id, 同 id 数据不会覆盖 -name: Deepseek-V3-VolcEngine -meta: - id: 65536 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - label_en: '' - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_basic.yaml b/backend/conf/model/template/entity/model_entity_template_basic.yaml deleted file mode 100644 index 33b21a7e..00000000 --- a/backend/conf/model/template/entity/model_entity_template_basic.yaml +++ /dev/null @@ -1,132 +0,0 @@ -id: 100 # 模型 entity id, 同 id 数据不会覆盖 -name: test_model -description: test_description -meta: - id: 0 -scenario: 1 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: frequency_penalty - label: - zh: 重复语句惩罚 - en: Frequency penalty - desc: - zh: '- **frequency penalty**: 当该值为正时，会阻止模型频繁使用相同的词汇和短语，从而增加输出内容的多样性。' - en: '**Frequency Penalty**: When positive, it discourages the model from repeating the same words and phrases, thereby increasing the diversity of the output.' - type: float - min: '-2' - max: '2' - precision: 2 - default_val: - default_val: '0' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: presence_penalty - label: - zh: 重复主题惩罚 - en: Presence penalty - desc: - zh: '- **presence penalty**: 当该值为正时，会阻止模型频繁讨论相同的主题，从而增加输出内容的多样性' - en: '**Presence Penalty**: When positive, it prevents the model from discussing the same topics repeatedly, thereby increasing the diversity of the output.' - type: float - min: '-2' - max: '2' - precision: 2 - default_val: - default_val: '0' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: response_format - label: - zh: 输出格式 - en: Response format - desc: - zh: '- **文本**: 使用普通文本格式回复\n- **Markdown**: 将引导模型使用Markdown格式输出回复\n- **JSON**: 将引导模型使用JSON格式输出' - en: '**Response Format**:\n\n- **Text**: Replies in plain text format\n- **Markdown**: Uses Markdown format for replies\n- **JSON**: Uses JSON format for replies' - type: int - precision: 0 - default_val: - default_val: '0' - options: - - - value: '0' - label: 'Text' - - - value: '1' - label: 'Markdown' - - - value: '2' - label: 'JSON' - style: - widget: radio_buttons - label: - zh: 输入及输出设置 - en: Input and output settings \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_claude.yaml b/backend/conf/model/template/entity/model_entity_template_claude.yaml deleted file mode 100644 index 41831168..00000000 --- a/backend/conf/model/template/entity/model_entity_template_claude.yaml +++ /dev/null @@ -1,47 +0,0 @@ -id: 2006 # 模型 entity id, 同 id 数据不会覆盖 -name: Claude-3.5-Sonnet -description: test claude description -meta: - id: 106 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings diff --git a/backend/conf/model/template/entity/model_entity_template_deepseek.yaml b/backend/conf/model/template/entity/model_entity_template_deepseek.yaml deleted file mode 100644 index f932c1fe..00000000 --- a/backend/conf/model/template/entity/model_entity_template_deepseek.yaml +++ /dev/null @@ -1,72 +0,0 @@ -id: 2004 # 模型 entity id, 同 id 数据不会覆盖 -name: DeepSeek-V3 -description: test deepseek description -meta: - id: 104 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成随机性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: response_format - label: - zh: 输出格式 - en: Response format - desc: - zh: '- **文本**: 使用普通文本格式回复\n- **JSON**: 将引导模型使用JSON格式输出' - en: '**Response Format**:\n\n- **Text**: Replies in plain text format\n- **Markdown**: Uses Markdown format for replies\n- **JSON**: Uses JSON format for replies' - type: int - precision: 0 - default_val: - default_val: '0' - options: - - - value: '0' - label: 'Text' - - - value: '1' - label: 'JSON Object' - style: - widget: radio_buttons - label: - zh: 输入及输出设置 - en: Input and output settings - diff --git a/backend/conf/model/template/entity/model_entity_template_gemini.yaml b/backend/conf/model/template/entity/model_entity_template_gemini.yaml deleted file mode 100644 index 3f04bed1..00000000 --- a/backend/conf/model/template/entity/model_entity_template_gemini.yaml +++ /dev/null @@ -1,90 +0,0 @@ -id: 2007 # 模型 entity id, 同 id 数据不会覆盖 -name: Gemini-2.5-Flash -description: test gemini description -meta: - id: 107 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: response_format - label: - zh: 输出格式 - en: Response format - desc: - zh: '- **文本**: 使用普通文本格式回复\n- **JSON**: 将引导模型使用JSON格式输出' - en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' - type: int - precision: 0 - default_val: - default_val: '0' - options: - - - value: '0' - label: 'Text' - - - value: '2' - label: 'JSON' - style: - widget: radio_buttons - label: - zh: 输入及输出设置 - en: Input and output settings \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_ollama.yaml b/backend/conf/model/template/entity/model_entity_template_ollama.yaml deleted file mode 100644 index cab168b4..00000000 --- a/backend/conf/model/template/entity/model_entity_template_ollama.yaml +++ /dev/null @@ -1,47 +0,0 @@ -id: 2003 # 模型 entity id, 同 id 数据不会覆盖 -name: Gemma-3 -description: test gemma-3 description -meta: - id: 103 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings diff --git a/backend/conf/model/template/entity/model_entity_template_openai.yaml b/backend/conf/model/template/entity/model_entity_template_openai.yaml deleted file mode 100644 index d7383e23..00000000 --- a/backend/conf/model/template/entity/model_entity_template_openai.yaml +++ /dev/null @@ -1,131 +0,0 @@ -id: 2001 # 模型 entity id, 同 id 数据不会覆盖 -name: GPT-4o -description: test gpt-4o description -meta: - id: 101 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.7' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: frequency_penalty - label: - zh: 重复语句惩罚 - en: Frequency penalty - desc: - zh: '- **frequency penalty**: 当该值为正时，会阻止模型频繁使用相同的词汇和短语，从而增加输出内容的多样性。' - en: '**Frequency Penalty**: When positive, it discourages the model from repeating the same words and phrases, thereby increasing the diversity of the output.' - type: float - min: '-2' - max: '2' - precision: 2 - default_val: - default_val: '0' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: presence_penalty - label: - zh: 重复主题惩罚 - en: Presence penalty - desc: - zh: '- **presence penalty**: 当该值为正时，会阻止模型频繁讨论相同的主题，从而增加输出内容的多样性' - en: '**Presence Penalty**: When positive, it prevents the model from discussing the same topics repeatedly, thereby increasing the diversity of the output.' - type: float - min: '-2' - max: '2' - precision: 2 - default_val: - default_val: '0' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: response_format - label: - zh: 输出格式 - en: Response format - desc: - zh: '- **文本**: 使用普通文本格式回复\n- **Markdown**: 将引导模型使用Markdown格式输出回复\n- **JSON**: 将引导模型使用JSON格式输出' - en: '**Response Format**:\n\n- **Text**: Replies in plain text format\n- **Markdown**: Uses Markdown format for replies\n- **JSON**: Uses JSON format for replies' - type: int - precision: 0 - default_val: - default_val: '0' - options: - - - value: '0' - label: 'Text' - - - value: '1' - label: 'Markdown' - - - value: '2' - label: 'JSON' - style: - widget: radio_buttons - label: - zh: 输入及输出设置 - en: Input and output settings \ No newline at end of file diff --git a/backend/conf/model/template/entity/model_entity_template_qwen.yaml b/backend/conf/model/template/entity/model_entity_template_qwen.yaml deleted file mode 100644 index bf9cb25e..00000000 --- a/backend/conf/model/template/entity/model_entity_template_qwen.yaml +++ /dev/null @@ -1,66 +0,0 @@ -id: 2005 # 模型 entity id, 同 id 数据不会覆盖 -name: Qwen3-32B -description: test qwen description -meta: - id: 105 -default_parameters: - - - name: temperature - label: - zh: 生成随机性 - en: Temperature - desc: - zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' - en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' - type: float - min: '0' - max: '1' - precision: 1 - default_val: - default_val: '1.0' - creative: '1' - balance: '0.8' - precise: '0.3' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity - - - name: max_tokens - label: - zh: 最大回复长度 - en: Response max length - desc: - zh: '控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。' - en: 'You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.' - type: int - min: '1' - max: '4096' - precision: 0 - default_val: - default_val: '4096' - style: - widget: slider - label: - zh: 输入及输出设置 - en: Input and output settings - - - name: top_p - label: - zh: Top P - en: Top P - desc: - zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' - en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' - type: float - min: '0' - max: '1' - precision: 2 - default_val: - default_val: '0.95' - style: - widget: slider - label: - zh: 生成多样性 - en: Generation diversity diff --git a/backend/conf/model/template/meta/model_meta_template_ark.yaml b/backend/conf/model/template/meta/model_meta_template_ark.yaml deleted file mode 100644 index 6d911c00..00000000 --- a/backend/conf/model/template/meta/model_meta_template_ark.yaml +++ /dev/null @@ -1,41 +0,0 @@ -# ark model template -# model list: https://www.volcengine.com/docs/82379/1330310 -# get api_key: https://www.volcengine.com/docs/82379/1399008#b00dee71 -# get region: https://www.volcengine.com/docs/82379/1319853#%E8%AE%BE%E7%BD%AE%E5%9C%B0%E5%9F%9F%E5%92%8C%E8%AE%BF%E9%97%AE%E5%9F%9F%E5%90%8D -# get ak/sk: https://www.volcengine.com/docs/82379/1319853#%E4%BD%BF%E7%94%A8access-key%E9%89%B4%E6%9D%83 -id: 102 # 模型 id, 同 id 数据不会覆盖 -name: Doubao # 模型 meta 名称 -icon_uri: doubao_v2.png -description: - zh: 豆包模型简介 # 模型默认描述 - en: doubao model description -protocol: ark # 模型连接协议 -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - - image - input_tokens: 128000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 16384 # 输出 token 上限 - max_tokens: 128000 # 最大 token 数量 - json_mode: false # 是否支持 json mode - prefix_caching: false # 是否支持 prefix caching - reasoning: false # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model name - temperature: 0.1 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 0.7 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - ark: # OPTIONAL - region: '' - access_key: '' - secret_key: '' - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-lite.yaml b/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-lite.yaml deleted file mode 100644 index d5f8ea14..00000000 --- a/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-lite.yaml +++ /dev/null @@ -1,40 +0,0 @@ -# ark model template -# model list: https://www.volcengine.com/docs/82379/1330310 -# get api_key: https://www.volcengine.com/docs/82379/1399008#b00dee71 -# get region: https://www.volcengine.com/docs/82379/1319853#%E8%AE%BE%E7%BD%AE%E5%9C%B0%E5%9F%9F%E5%92%8C%E8%AE%BF%E9%97%AE%E5%9F%9F%E5%90%8D -# get ak/sk: https://www.volcengine.com/docs/82379/1319853#%E4%BD%BF%E7%94%A8access-key%E9%89%B4%E6%9D%83 -id: 65536 # 模型 id, 同 id 数据不会覆盖 -name: doubao-1.5-lite # 模型 meta 名称 -icon_uri: doubao_v2.png -description: - zh: 'Doubao-1.5-lite，全新一代轻量版模型，极致响应速度，效果与时延均达到全球一流水平。' # 模型默认描述 - en: 'Doubao-1.5-lite, the new generation lightweight model, delivers ultra-fast response speed with both performance and latency reaching world-class standards.' -protocol: ark # 模型连接协议 -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - input_tokens: 20000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 12000 # 输出 token 上限 - max_tokens: 32000 # 最大 token 数量 - json_mode: false # 是否支持 json mode - prefix_caching: true # 是否支持 prefix caching - reasoning: false # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model name - temperature: 0.1 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 0.7 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - ark: # OPTIONAL - region: '' - access_key: '' - secret_key: '' - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-pro-256k.yaml b/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-pro-256k.yaml deleted file mode 100644 index 5d42f6cd..00000000 --- a/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-pro-256k.yaml +++ /dev/null @@ -1,40 +0,0 @@ -# ark model template -# model list: https://www.volcengine.com/docs/82379/1330310 -# get api_key: https://www.volcengine.com/docs/82379/1399008#b00dee71 -# get region: https://www.volcengine.com/docs/82379/1319853#%E8%AE%BE%E7%BD%AE%E5%9C%B0%E5%9F%9F%E5%92%8C%E8%AE%BF%E9%97%AE%E5%9F%9F%E5%90%8D -# get ak/sk: https://www.volcengine.com/docs/82379/1319853#%E4%BD%BF%E7%94%A8access-key%E9%89%B4%E6%9D%83 -id: 65536 # 模型 id, 同 id 数据不会覆盖 -name: doubao-1.5-pro-256k # 模型 meta 名称 -icon_uri: doubao_v2.png -description: - zh: 'doubao-1.5-pro-256k，基于doubao-1.5-Pro全面升级版，整体效果大幅提升10%。更高性能、更大窗口、超高性价比，适用于更广泛的应用场景。' # 模型默认描述 - en: 'doubao-1.5-pro-256k is a fully upgraded version based on doubao-1.5-Pro, with an overall performance improvement of 10%. It offers higher performance, a larger context window, and exceptional cost-effectiveness, making it suitable for a wider range of application scenarios.' -protocol: ark # 模型连接协议 -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - input_tokens: 96000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 12000 # 输出 token 上限 - max_tokens: 256000 # 最大 token 数量 - json_mode: false # 是否支持 json mode - prefix_caching: false # 是否支持 prefix caching - reasoning: false # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model name - temperature: 0.1 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 0.7 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - ark: # OPTIONAL - region: '' - access_key: '' - secret_key: '' - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-pro-32k.yaml b/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-pro-32k.yaml deleted file mode 100644 index 8da563a5..00000000 --- a/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-pro-32k.yaml +++ /dev/null @@ -1,40 +0,0 @@ -# ark model template -# model list: https://www.volcengine.com/docs/82379/1330310 -# get api_key: https://www.volcengine.com/docs/82379/1399008#b00dee71 -# get region: https://www.volcengine.com/docs/82379/1319853#%E8%AE%BE%E7%BD%AE%E5%9C%B0%E5%9F%9F%E5%92%8C%E8%AE%BF%E9%97%AE%E5%9F%9F%E5%90%8D -# get ak/sk: https://www.volcengine.com/docs/82379/1319853#%E4%BD%BF%E7%94%A8access-key%E9%89%B4%E6%9D%83 -id: 65536 # 模型 id, 同 id 数据不会覆盖 -name: doubao-1.5-pro-32k # 模型 meta 名称 -icon_uri: doubao_v2.png -description: - zh: 'Doubao-1.5-pro，全新一代主力模型，性能全面升级，在知识、代码、推理、等方面表现卓越。' # 模型默认描述 - en: 'Doubao-1.5-pro, the new generation flagship model, features comprehensive performance upgrades and excels in areas such as knowledge, coding, and reasoning.' -protocol: ark # 模型连接协议 -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - input_tokens: 96000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 12000 # 输出 token 上限 - max_tokens: 128000 # 最大 token 数量 - json_mode: false # 是否支持 json mode - prefix_caching: true # 是否支持 prefix caching - reasoning: false # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model name - temperature: 0.1 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 0.7 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - ark: # OPTIONAL - region: '' - access_key: '' - secret_key: '' - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-thinking-pro.yaml b/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-thinking-pro.yaml deleted file mode 100644 index 2e34c7d9..00000000 --- a/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-thinking-pro.yaml +++ /dev/null @@ -1,41 +0,0 @@ -# ark model template -# model list: https://www.volcengine.com/docs/82379/1330310 -# get api_key: https://www.volcengine.com/docs/82379/1399008#b00dee71 -# get region: https://www.volcengine.com/docs/82379/1319853#%E8%AE%BE%E7%BD%AE%E5%9C%B0%E5%9F%9F%E5%92%8C%E8%AE%BF%E9%97%AE%E5%9F%9F%E5%90%8D -# get ak/sk: https://www.volcengine.com/docs/82379/1319853#%E4%BD%BF%E7%94%A8access-key%E9%89%B4%E6%9D%83 -id: 65536 # 模型 id, 同 id 数据不会覆盖 -name: doubao-1.5-thinking-pro # 模型 meta 名称 -icon_uri: doubao_v2.png -description: - zh: 'doubao-1.5 全新深度思考模型，在数学、编程、科学推理等专业领域及创意写作等通用任务中表现突出，在 AIME 2024、Codeforces、GPQA 等多项权威基准上达到或接近业界第一梯队水平。' # 模型默认描述 - en: "doubao-1.5 is a brand-new deep thinking model that excels in specialized fields such as mathematics, programming, scientific reasoning, as well as general tasks like creative writing. It achieves or approaches the industry’s top-tier level on multiple authoritative benchmarks including AIME 2024, Codeforces, and GPQA." -protocol: ark # 模型连接协议 -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - - image - input_tokens: 96000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 16000 # 输出 token 上限 - max_tokens: 128000 # 最大 token 数量 - json_mode: true # 是否支持 json mode - prefix_caching: false # 是否支持 prefix caching - reasoning: true # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model name - temperature: 0.1 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 0.7 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - ark: # OPTIONAL - region: '' - access_key: '' - secret_key: '' - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-thinking-vision-pro.yaml b/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-thinking-vision-pro.yaml deleted file mode 100644 index 40748993..00000000 --- a/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-thinking-vision-pro.yaml +++ /dev/null @@ -1,42 +0,0 @@ -# ark model template -# model list: https://www.volcengine.com/docs/82379/1330310 -# get api_key: https://www.volcengine.com/docs/82379/1399008#b00dee71 -# get region: https://www.volcengine.com/docs/82379/1319853#%E8%AE%BE%E7%BD%AE%E5%9C%B0%E5%9F%9F%E5%92%8C%E8%AE%BF%E9%97%AE%E5%9F%9F%E5%90%8D -# get ak/sk: https://www.volcengine.com/docs/82379/1319853#%E4%BD%BF%E7%94%A8access-key%E9%89%B4%E6%9D%83 -id: 65536 # 模型 id, 同 id 数据不会覆盖 -name: doubao-1.5-thinking-vision-pro # 模型 meta 名称 -icon_uri: doubao_v2.png -description: - zh: 'doubao-1-5-thinking-vision-pro 全新视觉深度思考模型，具备更强的通用多模态理解和推理能力，在 59 个公开评测基准中的 37 个上取得 SOTA 表现。' # 模型默认描述 - en: 'doubao-1-5-thinking-vision-pro is a brand-new visual deep thinking model, featuring stronger general multimodal understanding and reasoning abilities, achieving SOTA performance on 37 out of 59 public evaluation benchmarks.' -protocol: ark # 模型连接协议 -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - - image - - video - input_tokens: 96000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 16000 # 输出 token 上限 - max_tokens: 128000 # 最大 token 数量 - json_mode: true # 是否支持 json mode - prefix_caching: false # 是否支持 prefix caching - reasoning: true # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model name - temperature: 0.1 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 0.7 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - ark: # OPTIONAL - region: '' - access_key: '' - secret_key: '' - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-vision-lite.yaml b/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-vision-lite.yaml deleted file mode 100644 index 5731fa84..00000000 --- a/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-vision-lite.yaml +++ /dev/null @@ -1,41 +0,0 @@ -# ark model template -# model list: https://www.volcengine.com/docs/82379/1330310 -# get api_key: https://www.volcengine.com/docs/82379/1399008#b00dee71 -# get region: https://www.volcengine.com/docs/82379/1319853#%E8%AE%BE%E7%BD%AE%E5%9C%B0%E5%9F%9F%E5%92%8C%E8%AE%BF%E9%97%AE%E5%9F%9F%E5%90%8D -# get ak/sk: https://www.volcengine.com/docs/82379/1319853#%E4%BD%BF%E7%94%A8access-key%E9%89%B4%E6%9D%83 -id: 65536 # 模型 id, 同 id 数据不会覆盖 -name: doubao-1.5-vision-lite # 模型 meta 名称 -icon_uri: doubao_v2.png -description: - zh: 'doubao-1.5-vision-lite，极具性价比的多模态大模型，支持任意分辨率和极端长宽比图像识别，增强视觉推理、文档识别、细节信息理解和指令遵循能力。' # 模型默认描述 - en: 'doubao-1.5-vision-lite is a highly cost-effective multimodal large model that supports image recognition at any resolution and extreme aspect ratios, enhancing visual reasoning, document recognition, detailed information comprehension, and instruction-following capabilities.' -protocol: ark # 模型连接协议 -capability: # 模型基础能力 - function_call: false # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - - image - input_tokens: 96000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 16000 # 输出 token 上限 - max_tokens: 128000 # 最大 token 数量 - json_mode: false # 是否支持 json mode - prefix_caching: false # 是否支持 prefix caching - reasoning: false # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model name - temperature: 0.1 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 0.7 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - ark: # OPTIONAL - region: '' - access_key: '' - secret_key: '' - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-vision-pro.yaml b/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-vision-pro.yaml deleted file mode 100644 index 807b4608..00000000 --- a/backend/conf/model/template/meta/model_meta_template_ark_doubao-1.5-vision-pro.yaml +++ /dev/null @@ -1,42 +0,0 @@ -# ark model template -# model list: https://www.volcengine.com/docs/82379/1330310 -# get api_key: https://www.volcengine.com/docs/82379/1399008#b00dee71 -# get region: https://www.volcengine.com/docs/82379/1319853#%E8%AE%BE%E7%BD%AE%E5%9C%B0%E5%9F%9F%E5%92%8C%E8%AE%BF%E9%97%AE%E5%9F%9F%E5%90%8D -# get ak/sk: https://www.volcengine.com/docs/82379/1319853#%E4%BD%BF%E7%94%A8access-key%E9%89%B4%E6%9D%83 -id: 65536 # 模型 id, 同 id 数据不会覆盖 -name: doubao-1.5-vision-pro # 模型 meta 名称 -icon_uri: doubao_v2.png -description: - zh: 'doubao-1.5-vision-pro，全新升级的多模态大模型，支持任意分辨率和极端长宽比图像识别，增强视觉推理、文档识别、细节信息理解和指令遵循能力。' # 模型默认描述 - en: 'doubao-1.5-vision-pro is a newly upgraded multimodal large model that supports image recognition at any resolution and extreme aspect ratios, enhancing visual reasoning, document recognition, detailed information comprehension, and instruction-following capabilities.' -protocol: ark # 模型连接协议 -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - - image - - video - input_tokens: 96000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 16000 # 输出 token 上限 - max_tokens: 128000 # 最大 token 数量 - json_mode: true # 是否支持 json mode - prefix_caching: false # 是否支持 prefix caching - reasoning: true # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model name - temperature: 0.1 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 0.7 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - ark: # OPTIONAL - region: '' - access_key: '' - secret_key: '' - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_ark_doubao-seed-1.6-flash.yaml b/backend/conf/model/template/meta/model_meta_template_ark_doubao-seed-1.6-flash.yaml deleted file mode 100644 index b8b11ff5..00000000 --- a/backend/conf/model/template/meta/model_meta_template_ark_doubao-seed-1.6-flash.yaml +++ /dev/null @@ -1,42 +0,0 @@ -# ark model template -# model list: https://www.volcengine.com/docs/82379/1330310 -# get api_key: https://www.volcengine.com/docs/82379/1399008#b00dee71 -# get region: https://www.volcengine.com/docs/82379/1319853#%E8%AE%BE%E7%BD%AE%E5%9C%B0%E5%9F%9F%E5%92%8C%E8%AE%BF%E9%97%AE%E5%9F%9F%E5%90%8D -# get ak/sk: https://www.volcengine.com/docs/82379/1319853#%E4%BD%BF%E7%94%A8access-key%E9%89%B4%E6%9D%83 -id: 65536 # 模型 id, 同 id 数据不会覆盖 -name: doubao-seed-1.6-flash # 模型 meta 名称 -icon_uri: doubao_v2.png -description: - zh: '有极致推理速度的多模态深度思考模型；同时支持文本和视觉理解。文本理解能力超过上一代 Lite 系列模型，视觉理解比肩友商 Pro 系列模型。' # 模型默认描述 - en: 'A multimodal deep thinking model with extreme reasoning speed; it supports both text and visual understanding. Its text comprehension surpasses the previous generation Lite series models, while its visual understanding rivals competitor Pro series models.' -protocol: ark # 模型连接协议 -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - - image - - video - input_tokens: 224000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 32000 # 输出 token 上限 - max_tokens: 256000 # 最大 token 数量 - json_mode: true # 是否支持 json mode - prefix_caching: true # 是否支持 prefix caching - reasoning: true # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model name - temperature: 0.1 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 0.7 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - ark: # OPTIONAL - region: '' - access_key: '' - secret_key: '' - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_ark_doubao-seed-1.6-thinking.yaml b/backend/conf/model/template/meta/model_meta_template_ark_doubao-seed-1.6-thinking.yaml deleted file mode 100644 index 217b8d6c..00000000 --- a/backend/conf/model/template/meta/model_meta_template_ark_doubao-seed-1.6-thinking.yaml +++ /dev/null @@ -1,42 +0,0 @@ -# ark model template -# model list: https://www.volcengine.com/docs/82379/1330310 -# get api_key: https://www.volcengine.com/docs/82379/1399008#b00dee71 -# get region: https://www.volcengine.com/docs/82379/1319853#%E8%AE%BE%E7%BD%AE%E5%9C%B0%E5%9F%9F%E5%92%8C%E8%AE%BF%E9%97%AE%E5%9F%9F%E5%90%8D -# get ak/sk: https://www.volcengine.com/docs/82379/1319853#%E4%BD%BF%E7%94%A8access-key%E9%89%B4%E6%9D%83 -id: 65536 # 模型 id, 同 id 数据不会覆盖 -name: doubao-seed-1.6-thinking # 模型 meta 名称 -icon_uri: doubao_v2.png -description: - zh: '在思考能力上进行了大幅强化， 对比 doubao 1.5 代深度理解模型，在编程、数学、逻辑推理等基础能力上进一步提升， 支持视觉理解。' # 模型默认描述 - en: 'Significantly enhanced in thinking capabilities, compared to the doubao 1.5 generation deep understanding model, with further improvements in fundamental skills such as programming, mathematics, and logical reasoning, and support for visual understanding.' -protocol: ark # 模型连接协议 -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - - image - - video - input_tokens: 224000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 16000 # 输出 token 上限 - max_tokens: 256000 # 最大 token 数量 - json_mode: true # 是否支持 json mode - prefix_caching: true # 是否支持 prefix caching - reasoning: true # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model name - temperature: 0.1 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 0.7 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - ark: # OPTIONAL - region: '' - access_key: '' - secret_key: '' - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_ark_doubao-seed-1.6.yaml b/backend/conf/model/template/meta/model_meta_template_ark_doubao-seed-1.6.yaml deleted file mode 100644 index a453d716..00000000 --- a/backend/conf/model/template/meta/model_meta_template_ark_doubao-seed-1.6.yaml +++ /dev/null @@ -1,42 +0,0 @@ -# ark model template -# model list: https://www.volcengine.com/docs/82379/1330310 -# get api_key: https://www.volcengine.com/docs/82379/1399008#b00dee71 -# get region: https://www.volcengine.com/docs/82379/1319853#%E8%AE%BE%E7%BD%AE%E5%9C%B0%E5%9F%9F%E5%92%8C%E8%AE%BF%E9%97%AE%E5%9F%9F%E5%90%8D -# get ak/sk: https://www.volcengine.com/docs/82379/1319853#%E4%BD%BF%E7%94%A8access-key%E9%89%B4%E6%9D%83 -id: 65536 # 模型 id, 同 id 数据不会覆盖 -name: doubao-seed-1.6 # 模型 meta 名称 -icon_uri: doubao_v2.png -description: - zh: '全新多模态深度思考模型，同时支持 thinking、non-thinking、auto三种思考模式。其中 non-thinking 模型对比 doubao-1.5-pro-32k-250115 模型大幅提升。' # 模型默认描述 - en: 'A brand-new multimodal deep thinking model supports three thinking modes: thinking, non-thinking, and auto. Among them, the non-thinking model has significantly improved compared to the doubao-1.5-pro-32k-250115 model.' -protocol: ark # 模型连接协议 -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - - image - - video - input_tokens: 224000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 32000 # 输出 token 上限 - max_tokens: 256000 # 最大 token 数量 - json_mode: true # 是否支持 json mode - prefix_caching: true # 是否支持 prefix caching - reasoning: true # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model name - temperature: 0.1 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 0.7 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - ark: # OPTIONAL - region: '' - access_key: '' - secret_key: '' - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_ark_volc_deepseek-r1.yaml b/backend/conf/model/template/meta/model_meta_template_ark_volc_deepseek-r1.yaml deleted file mode 100644 index eba59596..00000000 --- a/backend/conf/model/template/meta/model_meta_template_ark_volc_deepseek-r1.yaml +++ /dev/null @@ -1,40 +0,0 @@ -# ark model template -# model list: https://www.volcengine.com/docs/82379/1330310 -# get api_key: https://www.volcengine.com/docs/82379/1399008#b00dee71 -# get region: https://www.volcengine.com/docs/82379/1319853#%E8%AE%BE%E7%BD%AE%E5%9C%B0%E5%9F%9F%E5%92%8C%E8%AE%BF%E9%97%AE%E5%9F%9F%E5%90%8D -# get ak/sk: https://www.volcengine.com/docs/82379/1319853#%E4%BD%BF%E7%94%A8access-key%E9%89%B4%E6%9D%83 -id: 65536 # 模型 id, 同 id 数据不会覆盖 -name: deepseek-r1-ve # 模型 meta 名称 -icon_uri: deepseek_v2.png -description: - zh: 'deepseek-r1 是由深度求索推出的深度思考模型。在后训练阶段大规模使用了强化学习技术，在仅有极少标注数据的情况下，极大提升了模型推理能力。在数学、代码、自然语言推理等任务上，性能比肩 OpenAI o1 正式版。' # 模型默认描述 - en: "deepseek-r1 is a deep thinking model launched by Deep Seek. It extensively employs reinforcement learning during the post-training phase, significantly enhancing the model's reasoning ability with very limited annotated data. In tasks such as mathematics, coding, and natural language reasoning, its performance rivals that of the official OpenAI o1 version." -protocol: ark # 模型连接协议 -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - input_tokens: 96000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 32000 # 输出 token 上限 - max_tokens: 128000 # 最大 token 数量 - json_mode: false # 是否支持 json mode - prefix_caching: true # 是否支持 prefix caching - reasoning: true # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model name - temperature: 0.1 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 0.7 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - ark: # OPTIONAL - region: '' - access_key: '' - secret_key: '' - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_ark_volc_deepseek-v3.yaml b/backend/conf/model/template/meta/model_meta_template_ark_volc_deepseek-v3.yaml deleted file mode 100644 index 8b42724a..00000000 --- a/backend/conf/model/template/meta/model_meta_template_ark_volc_deepseek-v3.yaml +++ /dev/null @@ -1,40 +0,0 @@ -# ark model template -# model list: https://www.volcengine.com/docs/82379/1330310 -# get api_key: https://www.volcengine.com/docs/82379/1399008#b00dee71 -# get region: https://www.volcengine.com/docs/82379/1319853#%E8%AE%BE%E7%BD%AE%E5%9C%B0%E5%9F%9F%E5%92%8C%E8%AE%BF%E9%97%AE%E5%9F%9F%E5%90%8D -# get ak/sk: https://www.volcengine.com/docs/82379/1319853#%E4%BD%BF%E7%94%A8access-key%E9%89%B4%E6%9D%83 -id: 65536 # 模型 id, 同 id 数据不会覆盖 -name: deepseek-v3-ve # 模型 meta 名称 -icon_uri: deepseek_v2.png -description: - zh: 'deepseek-v3 由深度求索公司自研的MoE模型，多项评测成绩超越了 qwen2.5-72b 和 llama-3.1-405b 等开源模型，并在性能上和世界顶尖的闭源模型 gpt-4o 及 claude-3.5-Sonnet 不分伯仲。' # 模型默认描述 - en: "deepseek-v3 is a MoE model independently developed by Deep Seek. Its performance in multiple evaluations surpasses open-source models such as qwen2.5-72b and llama-3.1-405b, and it competes on par with world-leading closed-source models like gpt-4o and claude-3.5-Sonnet." -protocol: ark # 模型连接协议 -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - input_tokens: 96000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 16000 # 输出 token 上限 - max_tokens: 128000 # 最大 token 数量 - json_mode: false # 是否支持 json mode - prefix_caching: true # 是否支持 prefix caching - reasoning: false # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model name - temperature: 0.1 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 0.7 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - ark: # OPTIONAL - region: '' - access_key: '' - secret_key: '' - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_basic.yaml b/backend/conf/model/template/meta/model_meta_template_basic.yaml deleted file mode 100644 index 235fbcc2..00000000 --- a/backend/conf/model/template/meta/model_meta_template_basic.yaml +++ /dev/null @@ -1,75 +0,0 @@ -id: 0 # 模型 meta id, 同 id 数据不会覆盖 -name: test_model # 模型展示名称 -icon_uri: test_icon_uri # 模型展示图片 uri -icon_url: test_icon_url # 模型展示图片 url -description: - zh: test_description # 模型默认描述 - en: test_description -protocol: test_protocol # 模型连接协议, see: backend/infra/contract/chatmodel/protocol.go -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - - image - - audio - - video - input_tokens: 1024 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - - image - - audio - - video - output_tokens: 1024 # 输出 token 上限 - max_tokens: 2048 # 最大 token 数量 - json_mode: true # 是否支持 json mode - prefix_caching: false # 是否支持 prefix caching - reasoning: false # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - base_url: https://localhost:1234/chat/completion - api_key: qweasdzxc - timeout: 100 # nanosec - model: model_name # 模型名称 - temperature: 0.7 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 2048 # 默认 max_tokens - top_p: 0 # 默认 top_p - top_k: 0 # 默认 top_k - enable_thinking: false - stop: - - bye - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - openai: - by_azure: true - api_version: 2024-10-21 - response_format: - type: text - claude: - by_bedrock: true - access_key: bedrock_ak - secret_access_key: bedrock_secret_ak - session_token: bedrock_session_token - region: bedrock_region - ark: - region: region - access_key: ak - secret_key: sk - retry_times: 123 - custom_header: - key: val - deepseek: - response_format_type: text - gemini: - backend: 0 - project: '' - location: '' - api_version: '' - headers: - key_1: - - val_1 - - val_2 - timeout: 0 - include_thoughts: true - thinking_budget: null - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_claude.yaml b/backend/conf/model/template/meta/model_meta_template_claude.yaml deleted file mode 100644 index 6022cd58..00000000 --- a/backend/conf/model/template/meta/model_meta_template_claude.yaml +++ /dev/null @@ -1,40 +0,0 @@ -id: 106 # 模型 id, 同 id 数据不会覆盖 -name: Claude-3.5-Sonnet # 模型 meta 名称 -icon_uri: claude_v2.png # 模型展示图片 uri -icon_url: '' # 模型展示图片 url -description: - zh: claude 模型简介 # 模型默认描述 - en: claude model description -protocol: claude # 模型连接协议, see: backend/infra/contract/chatmodel/protocol.go -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - - image - input_tokens: 128000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 16384 # 输出 token 上限 - max_tokens: 128000 # 最大 token 数量 - json_mode: false # 是否支持 json mode - prefix_caching: false # 是否支持 prefix caching - reasoning: false # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - base_url: '' # REQUIRED: base_url - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model - temperature: 0.7 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 1 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - claude: # OPTIONAL - by_bedrock: false # true if using bedrock service - access_key: '' # access key - secret_access_key: '' # secret access key - session_token: '' # session_token - region: '' # region - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_deepseek.yaml b/backend/conf/model/template/meta/model_meta_template_deepseek.yaml deleted file mode 100644 index f655f7c7..00000000 --- a/backend/conf/model/template/meta/model_meta_template_deepseek.yaml +++ /dev/null @@ -1,35 +0,0 @@ -id: 104 # 模型 id, 同 id 数据不会覆盖 -name: DeepSeek-V3 # 模型 meta 名称 -icon_uri: deepseek_v2.png # 模型展示图片 uri -icon_url: '' # 模型展示图片 url -description: - zh: deepseek 模型简介 - en: deepseek model description -protocol: deepseek # 模型连接协议, see: backend/infra/contract/chatmodel/protocol.go -capability: # 模型基础能力 - function_call: false # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - input_tokens: 128000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 16384 # 输出 token 上限 - max_tokens: 128000 # 最大 token 数量 - json_mode: false # 是否支持 json mode - prefix_caching: false # 是否支持 prefix caching - reasoning: false # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - base_url: '' # REQUIRED: base_url - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model - temperature: 0.7 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 1 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - deepseek: # OPTIONAL - response_format_type: text # response format - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_gemini.yaml b/backend/conf/model/template/meta/model_meta_template_gemini.yaml deleted file mode 100644 index 672c3294..00000000 --- a/backend/conf/model/template/meta/model_meta_template_gemini.yaml +++ /dev/null @@ -1,48 +0,0 @@ -id: 107 # 模型 id, 同 id 数据不会覆盖 -name: Gemini-2.5-Flash # 模型 meta 名称 -icon_uri: gemini_v2.png # 模型展示图片 uri -icon_url: '' # 模型展示图片 url -description: - zh: gemini 模型简介 # 模型默认描述 - en: gemini model description -protocol: gemini # 模型连接协议, see: backend/infra/contract/chatmodel/protocol.go -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - - image - - audio - - video - input_tokens: 1048576 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 65536 # 输出 token 上限 - max_tokens: 1114112 # 最大 token 数量 - json_mode: true # 是否支持 json mode - prefix_caching: true # 是否支持 prefix caching - reasoning: true # 是否支持 reasoning - prefill_response: true # 是否支持续写 -conn_config: # 模型连接参数 - base_url: '' # REQUIRED: base_url - api_key: '' # REQUIRED: api_key - model: gemini-2.5-flash # REQUIRED: model - temperature: 0.7 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 1 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - gemini: - backend: 0 - project: '' - location: '' - api_version: '' - headers: - key_1: - - val_1 - - val_2 - timeout: 0 - include_thoughts: true - thinking_budget: null - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_ollama.yaml b/backend/conf/model/template/meta/model_meta_template_ollama.yaml deleted file mode 100644 index af462db1..00000000 --- a/backend/conf/model/template/meta/model_meta_template_ollama.yaml +++ /dev/null @@ -1,31 +0,0 @@ -id: 103 # 模型 id, 同 id 数据不会覆盖 -name: Gemma-3 # 模型 meta 名称 -icon_uri: ollama.png # 模型展示图片 uri -icon_url: '' # 模型展示图片 url -description: - zh: ollama 模型简介 - en: ollama model description -protocol: ollama # 模型连接协议, see: backend/infra/contract/chatmodel/protocol.go -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - input_tokens: 128000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 16384 # 输出 token 上限 - max_tokens: 128000 # 最大 token 数量 - json_mode: false # 是否支持 json mode - prefix_caching: false # 是否支持 prefix caching - reasoning: false # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - base_url: '' # REQUIRED: base_url - model: '' # REQUIRED: model - temperature: 0.6 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 0.95 # 默认 top_p - top_k: 20 # 默认 top_k - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_openai.yaml b/backend/conf/model/template/meta/model_meta_template_openai.yaml deleted file mode 100644 index 17d01289..00000000 --- a/backend/conf/model/template/meta/model_meta_template_openai.yaml +++ /dev/null @@ -1,39 +0,0 @@ -id: 101 # 模型 id, 同 id 数据不会覆盖 -name: GPT-4o # 模型 meta 名称 -icon_uri: openai_v2.png # 模型展示图片 uri -icon_url: '' # 模型展示图片 url -description: - zh: gpt 模型简介 - en: Multi-modal, 320ms, 88.7% MMLU, excels in education, customer support, health, and entertainment. # 模型默认描述 -protocol: openai # 模型连接协议, see: backend/infra/contract/chatmodel/protocol.go -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - - image - input_tokens: 128000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 16384 # 输出 token 上限 - max_tokens: 128000 # 最大 token 数量 - json_mode: false # 是否支持 json mode - prefix_caching: false # 是否支持 prefix caching - reasoning: false # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - base_url: '' # REQUIRED: base_url - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model - temperature: 0.7 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 1 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - openai: # OPTIONAL - by_azure: true # true if using azure openai - api_version: '' # azure api version - response_format: # response format - type: text - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/meta/model_meta_template_qwen.yaml b/backend/conf/model/template/meta/model_meta_template_qwen.yaml deleted file mode 100644 index 14531adb..00000000 --- a/backend/conf/model/template/meta/model_meta_template_qwen.yaml +++ /dev/null @@ -1,36 +0,0 @@ -id: 105 # 模型 id, 同 id 数据不会覆盖 -name: Qwen3-32B # 模型 meta 名称 -icon_uri: qwen_v2.png # 模型展示图片 uri -icon_url: '' # 模型展示图片 url -description: - zh: 通义千问模型 # 模型默认描述 - en: qwen model description -protocol: qwen # 模型连接协议, see: backend/infra/contract/chatmodel/protocol.go -capability: # 模型基础能力 - function_call: true # 模型是否支持 function call - input_modal: # 模型输入支持模态 - - text - input_tokens: 128000 # 输入 token 上限 - output_modal: # 模型输出支持模态 - - text - output_tokens: 16384 # 输出 token 上限 - max_tokens: 128000 # 最大 token 数量 - json_mode: false # 是否支持 json mode - prefix_caching: false # 是否支持 prefix caching - reasoning: false # 是否支持 reasoning - prefill_response: false # 是否支持续写 -conn_config: # 模型连接参数 - base_url: '' # REQUIRED: base_url - api_key: '' # REQUIRED: api_key - model: '' # REQUIRED: model - temperature: 0.7 # 默认 temperature - frequency_penalty: 0 # 默认 frequency_penalty - presence_penalty: 0 # 默认 presence_penalty - max_tokens: 4096 # 默认 max_tokens - top_p: 1 # 默认 top_p - top_k: 0 # 默认 top_k - # 下面是模型专用配置，仅配置 protocol 所对应的类型即可 - qwen: # OPTIONAL - response_format: # response format - type: text - status: 1 \ No newline at end of file diff --git a/backend/conf/model/template/model_template_ark.yaml b/backend/conf/model/template/model_template_ark.yaml new file mode 100755 index 00000000..d82584c8 --- /dev/null +++ b/backend/conf/model/template/model_template_ark.yaml @@ -0,0 +1,133 @@ +id: 2002 +name: Doubao Model +icon_uri: default_icon/doubao_v2.png +icon_url: "" +description: + zh: 豆包模型简介 + en: doubao model description +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: response_format + label: + zh: 输出格式 + en: Response format + desc: + zh: '- **文本**: 使用普通文本格式回复\n- **Markdown**: 将引导模型使用Markdown格式输出回复\n- **JSON**: 将引导模型使用JSON格式输出' + en: '**Response Format**:\n\n- **Text**: Replies in plain text format\n- **Markdown**: Uses Markdown format for replies\n- **JSON**: Uses JSON format for replies' + type: int + min: "" + max: "" + default_val: + default_val: "0" + options: + - label: Text + value: "0" + - label: Markdown + value: "1" + - label: JSON + value: "2" + style: + widget: radio_buttons + label: + zh: 输入及输出设置 + en: Input and output settings +meta: + name: Doubao + protocol: ark + capability: + function_call: true + input_modal: + - text + - image + input_tokens: 128000 + json_mode: false + max_tokens: 128000 + output_modal: + - text + output_tokens: 16384 + prefix_caching: false + reasoning: false + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.1 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 0.7 + top_k: 0 + stop: [] + openai: null + claude: null + ark: + region: "" + access_key: "" + secret_key: "" + retry_times: null + custom_header: {} + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_ark_doubao-1.5-lite.yaml b/backend/conf/model/template/model_template_ark_doubao-1.5-lite.yaml new file mode 100755 index 00000000..3040640d --- /dev/null +++ b/backend/conf/model/template/model_template_ark_doubao-1.5-lite.yaml @@ -0,0 +1,108 @@ +id: 65536 +name: Doubao-1.5-Lite +icon_uri: default_icon/doubao_v2.png +icon_url: "" +description: + zh: Doubao-1.5-lite，全新一代轻量版模型，极致响应速度，效果与时延均达到全球一流水平。 + en: Doubao-1.5-lite, the new generation lightweight model, delivers ultra-fast response speed with both performance and latency reaching world-class standards. +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity +meta: + name: doubao-1.5-lite + protocol: ark + capability: + function_call: true + input_modal: + - text + input_tokens: 20000 + json_mode: false + max_tokens: 32000 + output_modal: + - text + output_tokens: 12000 + prefix_caching: true + reasoning: false + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.1 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 0.7 + top_k: 0 + stop: [] + openai: null + claude: null + ark: + region: "" + access_key: "" + secret_key: "" + retry_times: null + custom_header: {} + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_ark_doubao-1.5-pro-256k.yaml b/backend/conf/model/template/model_template_ark_doubao-1.5-pro-256k.yaml new file mode 100755 index 00000000..e49e33ba --- /dev/null +++ b/backend/conf/model/template/model_template_ark_doubao-1.5-pro-256k.yaml @@ -0,0 +1,108 @@ +id: 65536 +name: Doubao-1.5-Pro-256k +icon_uri: default_icon/doubao_v2.png +icon_url: "" +description: + zh: doubao-1.5-pro-256k，基于doubao-1.5-Pro全面升级版，整体效果大幅提升10%。更高性能、更大窗口、超高性价比，适用于更广泛的应用场景。 + en: doubao-1.5-pro-256k is a fully upgraded version based on doubao-1.5-Pro, with an overall performance improvement of 10%. It offers higher performance, a larger context window, and exceptional cost-effectiveness, making it suitable for a wider range of application scenarios. +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity +meta: + name: doubao-1.5-pro-256k + protocol: ark + capability: + function_call: true + input_modal: + - text + input_tokens: 96000 + json_mode: false + max_tokens: 256000 + output_modal: + - text + output_tokens: 12000 + prefix_caching: false + reasoning: false + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.1 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 0.7 + top_k: 0 + stop: [] + openai: null + claude: null + ark: + region: "" + access_key: "" + secret_key: "" + retry_times: null + custom_header: {} + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_ark_doubao-1.5-pro-32k.yaml b/backend/conf/model/template/model_template_ark_doubao-1.5-pro-32k.yaml new file mode 100755 index 00000000..cc09ea65 --- /dev/null +++ b/backend/conf/model/template/model_template_ark_doubao-1.5-pro-32k.yaml @@ -0,0 +1,108 @@ +id: 65536 +name: Doubao-1.5-Pro-32k +icon_uri: default_icon/doubao_v2.png +icon_url: "" +description: + zh: Doubao-1.5-pro，全新一代主力模型，性能全面升级，在知识、代码、推理、等方面表现卓越。 + en: Doubao-1.5-pro, the new generation flagship model, features comprehensive performance upgrades and excels in areas such as knowledge, coding, and reasoning. +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity +meta: + name: doubao-1.5-pro-32k + protocol: ark + capability: + function_call: true + input_modal: + - text + input_tokens: 96000 + json_mode: false + max_tokens: 128000 + output_modal: + - text + output_tokens: 12000 + prefix_caching: true + reasoning: false + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.1 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 0.7 + top_k: 0 + stop: [] + openai: null + claude: null + ark: + region: "" + access_key: "" + secret_key: "" + retry_times: null + custom_header: {} + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_ark_doubao-1.5-thinking-pro.yaml b/backend/conf/model/template/model_template_ark_doubao-1.5-thinking-pro.yaml new file mode 100755 index 00000000..4eff7884 --- /dev/null +++ b/backend/conf/model/template/model_template_ark_doubao-1.5-thinking-pro.yaml @@ -0,0 +1,131 @@ +id: 65536 +name: Doubao-1.5-Thinking-Pro +icon_uri: default_icon/doubao_v2.png +icon_url: "" +description: + zh: doubao-1.5 全新深度思考模型，在数学、编程、科学推理等专业领域及创意写作等通用任务中表现突出，在 AIME 2024、Codeforces、GPQA 等多项权威基准上达到或接近业界第一梯队水平。 + en: doubao-1.5 is a brand-new deep thinking model that excels in specialized fields such as mathematics, programming, scientific reasoning, as well as general tasks like creative writing. It achieves or approaches the industry’s top-tier level on multiple authoritative benchmarks including AIME 2024, Codeforces, and GPQA. +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: response_format + label: + zh: 输出格式 + en: Response format + desc: + zh: '- **JSON**: 将引导模型使用JSON格式输出' + en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' + type: int + min: "" + max: "" + default_val: + default_val: "0" + options: + - label: Text + value: "0" + - label: JSON + value: "1" + style: + widget: radio_buttons + label: + zh: 输入及输出设置 + en: Input and output settings +meta: + name: doubao-1.5-thinking-pro + protocol: ark + capability: + function_call: true + input_modal: + - text + - image + input_tokens: 96000 + json_mode: true + max_tokens: 128000 + output_modal: + - text + output_tokens: 16000 + prefix_caching: false + reasoning: true + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.1 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 0.7 + top_k: 0 + stop: [] + openai: null + claude: null + ark: + region: "" + access_key: "" + secret_key: "" + retry_times: null + custom_header: {} + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_ark_doubao-1.5-thinking-vision-pro.yaml b/backend/conf/model/template/model_template_ark_doubao-1.5-thinking-vision-pro.yaml new file mode 100755 index 00000000..cf7e4c67 --- /dev/null +++ b/backend/conf/model/template/model_template_ark_doubao-1.5-thinking-vision-pro.yaml @@ -0,0 +1,132 @@ +id: 65536 +name: Doubao-1.5-Thinking-Vision-Pro +icon_uri: default_icon/doubao_v2.png +icon_url: "" +description: + zh: doubao-1-5-thinking-vision-pro 全新视觉深度思考模型，具备更强的通用多模态理解和推理能力，在 59 个公开评测基准中的 37 个上取得 SOTA 表现。 + en: doubao-1-5-thinking-vision-pro is a brand-new visual deep thinking model, featuring stronger general multimodal understanding and reasoning abilities, achieving SOTA performance on 37 out of 59 public evaluation benchmarks. +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: response_format + label: + zh: 输出格式 + en: Response format + desc: + zh: '- **JSON**: 将引导模型使用JSON格式输出' + en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' + type: int + min: "" + max: "" + default_val: + default_val: "0" + options: + - label: Text + value: "0" + - label: JSON + value: "1" + style: + widget: radio_buttons + label: + zh: 输入及输出设置 + en: Input and output settings +meta: + name: doubao-1.5-thinking-vision-pro + protocol: ark + capability: + function_call: true + input_modal: + - text + - image + - video + input_tokens: 96000 + json_mode: true + max_tokens: 128000 + output_modal: + - text + output_tokens: 16000 + prefix_caching: false + reasoning: true + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.1 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 0.7 + top_k: 0 + stop: [] + openai: null + claude: null + ark: + region: "" + access_key: "" + secret_key: "" + retry_times: null + custom_header: {} + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_ark_doubao-1.5-vision-lite.yaml b/backend/conf/model/template/model_template_ark_doubao-1.5-vision-lite.yaml new file mode 100755 index 00000000..abf737ca --- /dev/null +++ b/backend/conf/model/template/model_template_ark_doubao-1.5-vision-lite.yaml @@ -0,0 +1,131 @@ +id: 65536 +name: Doubao-1.5-Vision-Lite +icon_uri: default_icon/doubao_v2.png +icon_url: "" +description: + zh: doubao-1.5-vision-lite，极具性价比的多模态大模型，支持任意分辨率和极端长宽比图像识别，增强视觉推理、文档识别、细节信息理解和指令遵循能力。 + en: doubao-1.5-vision-lite is a highly cost-effective multimodal large model that supports image recognition at any resolution and extreme aspect ratios, enhancing visual reasoning, document recognition, detailed information comprehension, and instruction-following capabilities. +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: response_format + label: + zh: 输出格式 + en: Response format + desc: + zh: '- **JSON**: 将引导模型使用JSON格式输出' + en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' + type: int + min: "" + max: "" + default_val: + default_val: "0" + options: + - label: Text + value: "0" + - label: JSON + value: "1" + style: + widget: radio_buttons + label: + zh: 输入及输出设置 + en: Input and output settings +meta: + name: doubao-1.5-vision-lite + protocol: ark + capability: + function_call: false + input_modal: + - text + - image + input_tokens: 96000 + json_mode: false + max_tokens: 128000 + output_modal: + - text + output_tokens: 16000 + prefix_caching: false + reasoning: false + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.1 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 0.7 + top_k: 0 + stop: [] + openai: null + claude: null + ark: + region: "" + access_key: "" + secret_key: "" + retry_times: null + custom_header: {} + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_ark_doubao-1.5-vision-pro.yaml b/backend/conf/model/template/model_template_ark_doubao-1.5-vision-pro.yaml new file mode 100755 index 00000000..8c7c678d --- /dev/null +++ b/backend/conf/model/template/model_template_ark_doubao-1.5-vision-pro.yaml @@ -0,0 +1,132 @@ +id: 65536 +name: Doubao-1.5-Vision-Pro +icon_uri: default_icon/doubao_v2.png +icon_url: "" +description: + zh: doubao-1.5-vision-pro，全新升级的多模态大模型，支持任意分辨率和极端长宽比图像识别，增强视觉推理、文档识别、细节信息理解和指令遵循能力。 + en: doubao-1.5-vision-pro is a newly upgraded multimodal large model that supports image recognition at any resolution and extreme aspect ratios, enhancing visual reasoning, document recognition, detailed information comprehension, and instruction-following capabilities. +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: response_format + label: + zh: 输出格式 + en: Response format + desc: + zh: '- **JSON**: 将引导模型使用JSON格式输出' + en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' + type: int + min: "" + max: "" + default_val: + default_val: "0" + options: + - label: Text + value: "0" + - label: JSON + value: "1" + style: + widget: radio_buttons + label: + zh: 输入及输出设置 + en: Input and output settings +meta: + name: doubao-1.5-vision-pro + protocol: ark + capability: + function_call: true + input_modal: + - text + - image + - video + input_tokens: 96000 + json_mode: true + max_tokens: 128000 + output_modal: + - text + output_tokens: 16000 + prefix_caching: false + reasoning: true + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.1 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 0.7 + top_k: 0 + stop: [] + openai: null + claude: null + ark: + region: "" + access_key: "" + secret_key: "" + retry_times: null + custom_header: {} + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_ark_doubao-seed-1.6-flash.yaml b/backend/conf/model/template/model_template_ark_doubao-seed-1.6-flash.yaml new file mode 100755 index 00000000..6488f7be --- /dev/null +++ b/backend/conf/model/template/model_template_ark_doubao-seed-1.6-flash.yaml @@ -0,0 +1,132 @@ +id: 65536 +name: Doubao-Seed-1.6-Flash +icon_uri: default_icon/doubao_v2.png +icon_url: "" +description: + zh: 有极致推理速度的多模态深度思考模型；同时支持文本和视觉理解。文本理解能力超过上一代 Lite 系列模型，视觉理解比肩友商 Pro 系列模型。 + en: A multimodal deep thinking model with extreme reasoning speed; it supports both text and visual understanding. Its text comprehension surpasses the previous generation Lite series models, while its visual understanding rivals competitor Pro series models. +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: response_format + label: + zh: 输出格式 + en: Response format + desc: + zh: '- **JSON**: 将引导模型使用JSON格式输出' + en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' + type: int + min: "" + max: "" + default_val: + default_val: "0" + options: + - label: Text + value: "0" + - label: JSON + value: "1" + style: + widget: radio_buttons + label: + zh: 输入及输出设置 + en: Input and output settings +meta: + name: doubao-seed-1.6-flash + protocol: ark + capability: + function_call: true + input_modal: + - text + - image + - video + input_tokens: 224000 + json_mode: true + max_tokens: 256000 + output_modal: + - text + output_tokens: 32000 + prefix_caching: true + reasoning: true + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.1 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 0.7 + top_k: 0 + stop: [] + openai: null + claude: null + ark: + region: "" + access_key: "" + secret_key: "" + retry_times: null + custom_header: {} + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_ark_doubao-seed-1.6-thinking.yaml b/backend/conf/model/template/model_template_ark_doubao-seed-1.6-thinking.yaml new file mode 100755 index 00000000..c6d2f2e8 --- /dev/null +++ b/backend/conf/model/template/model_template_ark_doubao-seed-1.6-thinking.yaml @@ -0,0 +1,132 @@ +id: 65536 +name: Doubao-Seed-1.6-Thinking +icon_uri: default_icon/doubao_v2.png +icon_url: "" +description: + zh: 在思考能力上进行了大幅强化， 对比 doubao 1.5 代深度理解模型，在编程、数学、逻辑推理等基础能力上进一步提升， 支持视觉理解。 + en: Significantly enhanced in thinking capabilities, compared to the doubao 1.5 generation deep understanding model, with further improvements in fundamental skills such as programming, mathematics, and logical reasoning, and support for visual understanding. +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: response_format + label: + zh: 输出格式 + en: Response format + desc: + zh: '- **JSON**: 将引导模型使用JSON格式输出' + en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' + type: int + min: "" + max: "" + default_val: + default_val: "0" + options: + - label: Text + value: "0" + - label: JSON + value: "1" + style: + widget: radio_buttons + label: + zh: 输入及输出设置 + en: Input and output settings +meta: + name: doubao-seed-1.6-thinking + protocol: ark + capability: + function_call: true + input_modal: + - text + - image + - video + input_tokens: 224000 + json_mode: true + max_tokens: 256000 + output_modal: + - text + output_tokens: 16000 + prefix_caching: true + reasoning: true + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.1 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 0.7 + top_k: 0 + stop: [] + openai: null + claude: null + ark: + region: "" + access_key: "" + secret_key: "" + retry_times: null + custom_header: {} + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_ark_doubao-seed-1.6.yaml b/backend/conf/model/template/model_template_ark_doubao-seed-1.6.yaml new file mode 100755 index 00000000..1571dd7c --- /dev/null +++ b/backend/conf/model/template/model_template_ark_doubao-seed-1.6.yaml @@ -0,0 +1,132 @@ +id: 65536 +name: Doubao-Seed-1.6 +icon_uri: default_icon/doubao_v2.png +icon_url: "" +description: + zh: 全新多模态深度思考模型，同时支持 thinking、non-thinking、auto三种思考模式。其中 non-thinking 模型对比 doubao-1.5-pro-32k-250115 模型大幅提升。 + en: 'A brand-new multimodal deep thinking model supports three thinking modes: thinking, non-thinking, and auto. Among them, the non-thinking model has significantly improved compared to the doubao-1.5-pro-32k-250115 model.' +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: response_format + label: + zh: 输出格式 + en: Response format + desc: + zh: '- **JSON**: 将引导模型使用JSON格式输出' + en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' + type: int + min: "" + max: "" + default_val: + default_val: "0" + options: + - label: Text + value: "0" + - label: JSON + value: "1" + style: + widget: radio_buttons + label: + zh: 输入及输出设置 + en: Input and output settings +meta: + name: doubao-seed-1.6 + protocol: ark + capability: + function_call: true + input_modal: + - text + - image + - video + input_tokens: 224000 + json_mode: true + max_tokens: 256000 + output_modal: + - text + output_tokens: 32000 + prefix_caching: true + reasoning: true + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.1 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 0.7 + top_k: 0 + stop: [] + openai: null + claude: null + ark: + region: "" + access_key: "" + secret_key: "" + retry_times: null + custom_header: {} + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_ark_volc_deepseek-r1.yaml b/backend/conf/model/template/model_template_ark_volc_deepseek-r1.yaml new file mode 100755 index 00000000..ae72f3fa --- /dev/null +++ b/backend/conf/model/template/model_template_ark_volc_deepseek-r1.yaml @@ -0,0 +1,108 @@ +id: 65536 +name: Deepseek-R1-VolcEngine +icon_uri: default_icon/deepseek_v2.png +icon_url: "" +description: + zh: deepseek-r1 是由深度求索推出的深度思考模型。在后训练阶段大规模使用了强化学习技术，在仅有极少标注数据的情况下，极大提升了模型推理能力。在数学、代码、自然语言推理等任务上，性能比肩 OpenAI o1 正式版。 + en: deepseek-r1 is a deep thinking model launched by Deep Seek. It extensively employs reinforcement learning during the post-training phase, significantly enhancing the model's reasoning ability with very limited annotated data. In tasks such as mathematics, coding, and natural language reasoning, its performance rivals that of the official OpenAI o1 version. +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity +meta: + name: deepseek-r1-ve + protocol: ark + capability: + function_call: true + input_modal: + - text + input_tokens: 96000 + json_mode: false + max_tokens: 128000 + output_modal: + - text + output_tokens: 32000 + prefix_caching: true + reasoning: true + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.1 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 0.7 + top_k: 0 + stop: [] + openai: null + claude: null + ark: + region: "" + access_key: "" + secret_key: "" + retry_times: null + custom_header: {} + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_ark_volc_deepseek-v3.yaml b/backend/conf/model/template/model_template_ark_volc_deepseek-v3.yaml new file mode 100755 index 00000000..efa1fcef --- /dev/null +++ b/backend/conf/model/template/model_template_ark_volc_deepseek-v3.yaml @@ -0,0 +1,108 @@ +id: 65536 +name: Deepseek-V3-VolcEngine +icon_uri: default_icon/deepseek_v2.png +icon_url: "" +description: + zh: deepseek-v3 由深度求索公司自研的MoE模型，多项评测成绩超越了 qwen2.5-72b 和 llama-3.1-405b 等开源模型，并在性能上和世界顶尖的闭源模型 gpt-4o 及 claude-3.5-Sonnet 不分伯仲。 + en: deepseek-v3 is a MoE model independently developed by Deep Seek. Its performance in multiple evaluations surpasses open-source models such as qwen2.5-72b and llama-3.1-405b, and it competes on par with world-leading closed-source models like gpt-4o and claude-3.5-Sonnet. +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity +meta: + name: deepseek-v3-ve + protocol: ark + capability: + function_call: true + input_modal: + - text + input_tokens: 96000 + json_mode: false + max_tokens: 128000 + output_modal: + - text + output_tokens: 16000 + prefix_caching: true + reasoning: false + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.1 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 0.7 + top_k: 0 + stop: [] + openai: null + claude: null + ark: + region: "" + access_key: "" + secret_key: "" + retry_times: null + custom_header: {} + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_basic.yaml b/backend/conf/model/template/model_template_basic.yaml new file mode 100755 index 00000000..a46d09a3 --- /dev/null +++ b/backend/conf/model/template/model_template_basic.yaml @@ -0,0 +1,201 @@ +id: 100 +name: test_model +icon_uri: default_icon/test_icon_uri.png +icon_url: test_icon_url +description: + zh: test_description + en: test_description +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: frequency_penalty + label: + zh: 重复语句惩罚 + en: Frequency penalty + desc: + zh: '- **frequency penalty**: 当该值为正时，会阻止模型频繁使用相同的词汇和短语，从而增加输出内容的多样性。' + en: '**Frequency Penalty**: When positive, it discourages the model from repeating the same words and phrases, thereby increasing the diversity of the output.' + type: float + min: "-2" + max: "2" + default_val: + default_val: "0" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: presence_penalty + label: + zh: 重复主题惩罚 + en: Presence penalty + desc: + zh: '- **presence penalty**: 当该值为正时，会阻止模型频繁讨论相同的主题，从而增加输出内容的多样性' + en: '**Presence Penalty**: When positive, it prevents the model from discussing the same topics repeatedly, thereby increasing the diversity of the output.' + type: float + min: "-2" + max: "2" + default_val: + default_val: "0" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: response_format + label: + zh: 输出格式 + en: Response format + desc: + zh: '- **文本**: 使用普通文本格式回复\n- **Markdown**: 将引导模型使用Markdown格式输出回复\n- **JSON**: 将引导模型使用JSON格式输出' + en: '**Response Format**:\n\n- **Text**: Replies in plain text format\n- **Markdown**: Uses Markdown format for replies\n- **JSON**: Uses JSON format for replies' + type: int + min: "" + max: "" + default_val: + default_val: "0" + options: + - label: Text + value: "0" + - label: Markdown + value: "1" + - label: JSON + value: "2" + style: + widget: radio_buttons + label: + zh: 输入及输出设置 + en: Input and output settings +meta: + name: test_model + protocol: test_protocol + capability: + function_call: true + input_modal: + - text + - image + - audio + - video + input_tokens: 1024 + json_mode: true + max_tokens: 2048 + output_modal: + - text + - image + - audio + - video + output_tokens: 1024 + prefix_caching: false + reasoning: false + prefill_response: false + conn_config: + base_url: https://localhost:1234/chat/completion + api_key: qweasdzxc + timeout: 10s + model: model_name + temperature: 0.7 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 2048 + top_p: 0 + top_k: 0 + stop: + - bye + enable_thinking: false + openai: + by_azure: true + api_version: "2024-10-21" + response_format: + type: text + jsonschema: null + claude: + by_bedrock: true + access_key: bedrock_ak + secret_access_key: bedrock_secret_ak + session_token: bedrock_session_token + region: bedrock_region + ark: + region: region + access_key: ak + secret_key: sk + retry_times: 123 + custom_header: + key: val + deepseek: + response_format_type: text + qwen: null + gemini: + backend: 0 + project: "" + location: "" + api_version: "" + headers: + key_1: + - val_1 + - val_2 + timeout_ms: 0 + include_thoughts: true + thinking_budget: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_claude.yaml b/backend/conf/model/template/model_template_claude.yaml new file mode 100755 index 00000000..e3d0200e --- /dev/null +++ b/backend/conf/model/template/model_template_claude.yaml @@ -0,0 +1,90 @@ +id: 2006 +name: Claude-3.5-Sonnet +icon_uri: default_icon/claude_v2.png +icon_url: "" +description: + zh: claude 模型简介 + en: claude model description +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings +meta: + name: Claude-3.5-Sonnet + protocol: claude + capability: + function_call: true + input_modal: + - text + - image + input_tokens: 128000 + json_mode: false + max_tokens: 128000 + output_modal: + - text + output_tokens: 16384 + prefix_caching: false + reasoning: false + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.7 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 1 + top_k: 0 + stop: [] + openai: null + claude: + by_bedrock: false + access_key: "" + secret_access_key: "" + session_token: "" + region: "" + ark: null + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_deepseek.yaml b/backend/conf/model/template/model_template_deepseek.yaml new file mode 100755 index 00000000..8eef84d9 --- /dev/null +++ b/backend/conf/model/template/model_template_deepseek.yaml @@ -0,0 +1,107 @@ +id: 2004 +name: DeepSeek-V3 +icon_uri: default_icon/deepseek_v2.png +icon_url: "" +description: + zh: deepseek 模型简介 + en: deepseek model description +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成随机性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: response_format + label: + zh: 输出格式 + en: Response format + desc: + zh: '- **文本**: 使用普通文本格式回复\n- **JSON**: 将引导模型使用JSON格式输出' + en: '**Response Format**:\n\n- **Text**: Replies in plain text format\n- **Markdown**: Uses Markdown format for replies\n- **JSON**: Uses JSON format for replies' + type: int + min: "" + max: "" + default_val: + default_val: "0" + options: + - label: Text + value: "0" + - label: JSON Object + value: "1" + style: + widget: radio_buttons + label: + zh: 输入及输出设置 + en: Input and output settings +meta: + name: DeepSeek-V3 + protocol: deepseek + capability: + function_call: false + input_modal: + - text + input_tokens: 128000 + json_mode: false + max_tokens: 128000 + output_modal: + - text + output_tokens: 16384 + prefix_caching: false + reasoning: false + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.7 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 1 + top_k: 0 + stop: [] + openai: null + claude: null + ark: null + deepseek: + response_format_type: text + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_gemini.yaml b/backend/conf/model/template/model_template_gemini.yaml new file mode 100755 index 00000000..1eb2bd4c --- /dev/null +++ b/backend/conf/model/template/model_template_gemini.yaml @@ -0,0 +1,139 @@ +id: 2007 +name: Gemini-2.5-Flash +icon_uri: default_icon/gemini_v2.png +icon_url: "" +description: + zh: gemini 模型简介 + en: gemini model description +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: response_format + label: + zh: 输出格式 + en: Response format + desc: + zh: '- **文本**: 使用普通文本格式回复\n- **JSON**: 将引导模型使用JSON格式输出' + en: '**Response Format**:\n\n- **JSON**: Uses JSON format for replies' + type: int + min: "" + max: "" + default_val: + default_val: "0" + options: + - label: Text + value: "0" + - label: JSON + value: "2" + style: + widget: radio_buttons + label: + zh: 输入及输出设置 + en: Input and output settings +meta: + name: Gemini-2.5-Flash + protocol: gemini + capability: + function_call: true + input_modal: + - text + - image + - audio + - video + input_tokens: 1048576 + json_mode: true + max_tokens: 1114112 + output_modal: + - text + output_tokens: 65536 + prefix_caching: true + reasoning: true + prefill_response: true + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: gemini-2.5-flash + temperature: 0.7 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 1 + top_k: 0 + stop: [] + openai: null + claude: null + ark: null + deepseek: null + qwen: null + gemini: + backend: 0 + project: "" + location: "" + api_version: "" + headers: + key_1: + - val_1 + - val_2 + timeout_ms: 0 + include_thoughts: true + thinking_budget: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_ollama.yaml b/backend/conf/model/template/model_template_ollama.yaml new file mode 100755 index 00000000..63d5048f --- /dev/null +++ b/backend/conf/model/template/model_template_ollama.yaml @@ -0,0 +1,84 @@ +id: 2003 +name: Gemma-3 +icon_uri: default_icon/ollama.png +icon_url: "" +description: + zh: ollama 模型简介 + en: ollama model description +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings +meta: + name: Gemma-3 + protocol: ollama + capability: + function_call: true + input_modal: + - text + input_tokens: 128000 + json_mode: false + max_tokens: 128000 + output_modal: + - text + output_tokens: 16384 + prefix_caching: false + reasoning: false + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.6 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 0.95 + top_k: 20 + stop: [] + openai: null + claude: null + ark: null + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_openai.yaml b/backend/conf/model/template/model_template_openai.yaml new file mode 100755 index 00000000..5a32b075 --- /dev/null +++ b/backend/conf/model/template/model_template_openai.yaml @@ -0,0 +1,171 @@ +id: 2001 +name: GPT-4o +icon_uri: default_icon/openai_v2.png +icon_url: "" +description: + zh: gpt 模型简介 + en: Multi-modal, 320ms, 88.7% MMLU, excels in education, customer support, health, and entertainment. +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.7" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: frequency_penalty + label: + zh: 重复语句惩罚 + en: Frequency penalty + desc: + zh: '- **frequency penalty**: 当该值为正时，会阻止模型频繁使用相同的词汇和短语，从而增加输出内容的多样性。' + en: '**Frequency Penalty**: When positive, it discourages the model from repeating the same words and phrases, thereby increasing the diversity of the output.' + type: float + min: "-2" + max: "2" + default_val: + default_val: "0" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: presence_penalty + label: + zh: 重复主题惩罚 + en: Presence penalty + desc: + zh: '- **presence penalty**: 当该值为正时，会阻止模型频繁讨论相同的主题，从而增加输出内容的多样性' + en: '**Presence Penalty**: When positive, it prevents the model from discussing the same topics repeatedly, thereby increasing the diversity of the output.' + type: float + min: "-2" + max: "2" + default_val: + default_val: "0" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: response_format + label: + zh: 输出格式 + en: Response format + desc: + zh: '- **文本**: 使用普通文本格式回复\n- **Markdown**: 将引导模型使用Markdown格式输出回复\n- **JSON**: 将引导模型使用JSON格式输出' + en: '**Response Format**:\n\n- **Text**: Replies in plain text format\n- **Markdown**: Uses Markdown format for replies\n- **JSON**: Uses JSON format for replies' + type: int + min: "" + max: "" + default_val: + default_val: "0" + options: + - label: Text + value: "0" + - label: Markdown + value: "1" + - label: JSON + value: "2" + style: + widget: radio_buttons + label: + zh: 输入及输出设置 + en: Input and output settings +meta: + name: GPT-4o + protocol: openai + capability: + function_call: true + input_modal: + - text + - image + input_tokens: 128000 + json_mode: false + max_tokens: 128000 + output_modal: + - text + output_tokens: 16384 + prefix_caching: false + reasoning: false + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.7 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 1 + top_k: 0 + stop: [] + openai: + by_azure: true + api_version: "" + response_format: + type: text + jsonschema: null + claude: null + ark: null + deepseek: null + qwen: null + gemini: null + custom: {} + status: 0 diff --git a/backend/conf/model/template/model_template_qwen.yaml b/backend/conf/model/template/model_template_qwen.yaml new file mode 100755 index 00000000..6b56d846 --- /dev/null +++ b/backend/conf/model/template/model_template_qwen.yaml @@ -0,0 +1,106 @@ +id: 2005 +name: Qwen3-32B +icon_uri: default_icon/qwen_v2.png +icon_url: "" +description: + zh: 通义千问模型 + en: qwen model description +default_parameters: + - name: temperature + label: + zh: 生成随机性 + en: Temperature + desc: + zh: '- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。' + en: '**Temperature**:\n\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\n- It is recommended not to adjust this value with \"Top p\" at the same time.' + type: float + min: "0" + max: "1" + default_val: + balance: "0.8" + creative: "1" + default_val: "1.0" + precise: "0.3" + precision: 1 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity + - name: max_tokens + label: + zh: 最大回复长度 + en: Response max length + desc: + zh: 控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。 + en: You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters. + type: int + min: "1" + max: "4096" + default_val: + default_val: "4096" + options: [] + style: + widget: slider + label: + zh: 输入及输出设置 + en: Input and output settings + - name: top_p + label: + zh: Top P + en: Top P + desc: + zh: '- **Top p 为累计概率**: 模型在生成输出时会从概率最高的词汇开始选择，直到这些词汇的总概率累积达到Top p 值。这样可以限制模型只选择这些高概率的词汇，从而控制输出内容的多样性。建议不要与“生成随机性”同时调整。' + en: '**Top P**:\n\n- An alternative to sampling with temperature, where only tokens within the top p probability mass are considered. For example, 0.1 means only the top 10% probability mass tokens are considered.\n- We recommend altering this or temperature, but not both.' + type: float + min: "0" + max: "1" + default_val: + default_val: "0.95" + precision: 2 + options: [] + style: + widget: slider + label: + zh: 生成多样性 + en: Generation diversity +meta: + name: Qwen3-32B + protocol: qwen + capability: + function_call: true + input_modal: + - text + input_tokens: 128000 + json_mode: false + max_tokens: 128000 + output_modal: + - text + output_tokens: 16384 + prefix_caching: false + reasoning: false + prefill_response: false + conn_config: + base_url: "" + api_key: "" + timeout: 0s + model: "" + temperature: 0.7 + frequency_penalty: 0 + presence_penalty: 0 + max_tokens: 4096 + top_p: 1 + top_k: 0 + stop: [] + openai: null + claude: null + ark: null + deepseek: null + qwen: + response_format: + type: text + jsonschema: null + gemini: null + custom: {} + status: 0 diff --git a/backend/crossdomain/contract/crossmodelmgr/cross_modelmgr.go b/backend/crossdomain/contract/crossmodelmgr/cross_modelmgr.go deleted file mode 100644 index eab97c33..00000000 --- a/backend/crossdomain/contract/crossmodelmgr/cross_modelmgr.go +++ /dev/null @@ -1,39 +0,0 @@ -/* - * Copyright 2025 coze-dev Authors - * - * Licensed under the Apache License, Version 2.0 (the "License"); - * you may not use this file except in compliance with the License. - * You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package crossmodelmgr - -import ( - "context" - - "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" -) - -type ModelMgr interface { - MGetModelByID(ctx context.Context, req *modelmgr.MGetModelRequest) ([]*modelmgr.Model, error) -} - -type Model = modelmgr.Model - -var defaultSVC ModelMgr - -func DefaultSVC() ModelMgr { - return defaultSVC -} - -func SetDefaultSVC(c ModelMgr) { - defaultSVC = c -} diff --git a/backend/crossdomain/impl/modelmgr/modelmgr.go b/backend/crossdomain/impl/modelmgr/modelmgr.go deleted file mode 100644 index 8c49907a..00000000 --- a/backend/crossdomain/impl/modelmgr/modelmgr.go +++ /dev/null @@ -1,52 +0,0 @@ -/* - * Copyright 2025 coze-dev Authors - * - * Licensed under the Apache License, Version 2.0 (the "License"); - * you may not use this file except in compliance with the License. - * You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package modelmgr - -import ( - "context" - - model "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" - "github.com/coze-dev/coze-studio/backend/crossdomain/contract/crossmodelmgr" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr" -) - -var defaultSVC crossmodelmgr.ModelMgr - -type impl struct { - DomainSVC modelmgr.Manager -} - -func InitDomainService(c modelmgr.Manager) crossmodelmgr.ModelMgr { - defaultSVC = &impl{ - DomainSVC: c, - } - return defaultSVC -} - -func (s *impl) MGetModelByID(ctx context.Context, req *modelmgr.MGetModelRequest) ([]*model.Model, error) { - res, err := s.DomainSVC.MGetModelByID(ctx, req) - if err != nil { - return nil, err - } - - ret := make([]*model.Model, 0, len(res)) - for _, v := range res { - ret = append(ret, v.Model) - } - - return ret, nil -} diff --git a/backend/crossdomain/workflow/model/model.go b/backend/crossdomain/workflow/model/model.go index 7d19ca6b..df356dd9 100644 --- a/backend/crossdomain/workflow/model/model.go +++ b/backend/crossdomain/workflow/model/model.go @@ -22,10 +22,9 @@ import ( model2 "github.com/cloudwego/eino/components/model" - "github.com/coze-dev/coze-studio/backend/crossdomain/contract/crossmodelmgr" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr" "github.com/coze-dev/coze-studio/backend/domain/workflow/crossdomain/model" "github.com/coze-dev/coze-studio/backend/infra/contract/chatmodel" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" chatmodel2 "github.com/coze-dev/coze-studio/backend/infra/impl/chatmodel" "github.com/coze-dev/coze-studio/backend/pkg/lang/ptr" ) @@ -45,9 +44,9 @@ func NewModelManager(m modelmgr.Manager, f chatmodel.Factory) *ModelManager { } } -func (m *ModelManager) GetModel(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *crossmodelmgr.Model, error) { +func (m *ModelManager) GetModel(ctx context.Context, params *model.LLMParams) (model2.BaseChatModel, *modelmgr.Model, error) { modelID := params.ModelType - models, err := crossmodelmgr.DefaultSVC().MGetModelByID(ctx, &modelmgr.MGetModelRequest{ + models, err := m.modelMgr.MGetModelByID(ctx, &modelmgr.MGetModelRequest{ IDs: []int64{modelID}, }) if err != nil { @@ -55,7 +54,7 @@ func (m *ModelManager) GetModel(ctx context.Context, params *model.LLMParams) (m } var config *chatmodel.Config var protocol chatmodel.Protocol - var mdl *crossmodelmgr.Model + var mdl *modelmgr.Model for i := range models { md := models[i] if md.ID == modelID { diff --git a/backend/domain/agent/singleagent/internal/agentflow/agent_flow_builder.go b/backend/domain/agent/singleagent/internal/agentflow/agent_flow_builder.go index c0f73947..7a0e7126 100644 --- a/backend/domain/agent/singleagent/internal/agentflow/agent_flow_builder.go +++ b/backend/domain/agent/singleagent/internal/agentflow/agent_flow_builder.go @@ -26,6 +26,7 @@ import ( "github.com/cloudwego/eino/compose" "github.com/cloudwego/eino/flow/agent/react" "github.com/cloudwego/eino/schema" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" "github.com/coze-dev/coze-studio/backend/domain/agent/singleagent/entity" "github.com/coze-dev/coze-studio/backend/domain/workflow" @@ -38,6 +39,7 @@ type Config struct { Agent *entity.SingleAgent UserID string Identity *entity.AgentIdentity + ModelMgr modelmgr.Manager ModelFactory chatmodel.Factory CPStore compose.CheckPointStore } @@ -86,7 +88,7 @@ func BuildAgent(ctx context.Context, conf *Config) (r *AgentRunner, err error) { return nil, err } - modelInfo, err := loadModelInfo(ctx, ptr.From(conf.Agent.ModelInfo.ModelId)) + modelInfo, err := loadModelInfo(ctx, conf.ModelMgr, ptr.From(conf.Agent.ModelInfo.ModelId)) if err != nil { return nil, err } diff --git a/backend/domain/agent/singleagent/internal/agentflow/agent_flow_runner.go b/backend/domain/agent/singleagent/internal/agentflow/agent_flow_runner.go index 6bbfc022..c7574d8f 100644 --- a/backend/domain/agent/singleagent/internal/agentflow/agent_flow_runner.go +++ b/backend/domain/agent/singleagent/internal/agentflow/agent_flow_runner.go @@ -27,11 +27,10 @@ import ( "github.com/cloudwego/eino/schema" "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/agentrun" - "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/singleagent" - "github.com/coze-dev/coze-studio/backend/crossdomain/contract/crossmodelmgr" "github.com/coze-dev/coze-studio/backend/crossdomain/contract/crossworkflow" "github.com/coze-dev/coze-studio/backend/domain/agent/singleagent/entity" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" "github.com/coze-dev/coze-studio/backend/pkg/lang/conv" "github.com/coze-dev/coze-studio/backend/pkg/logs" ) @@ -59,7 +58,7 @@ type AgentRunner struct { requireCheckpoint bool containWfTool bool - modelInfo *crossmodelmgr.Model + modelInfo *modelmgr.Model } func (r *AgentRunner) StreamExecute(ctx context.Context, req *AgentRequest) ( diff --git a/backend/domain/agent/singleagent/internal/agentflow/node_chat_model.go b/backend/domain/agent/singleagent/internal/agentflow/node_chat_model.go index 48a316ef..efd479e6 100644 --- a/backend/domain/agent/singleagent/internal/agentflow/node_chat_model.go +++ b/backend/domain/agent/singleagent/internal/agentflow/node_chat_model.go @@ -20,16 +20,15 @@ import ( "context" "fmt" - "github.com/coze-dev/coze-studio/backend/crossdomain/contract/crossmodelmgr" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr" "github.com/coze-dev/coze-studio/backend/infra/contract/chatmodel" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" "github.com/coze-dev/coze-studio/backend/pkg/errorx" "github.com/coze-dev/coze-studio/backend/types/errno" ) type config struct { modelFactory chatmodel.Factory - modelInfo *crossmodelmgr.Model + modelInfo *modelmgr.Model } func newChatModel(ctx context.Context, conf *config) (chatmodel.ToolCallingChatModel, error) { @@ -53,12 +52,12 @@ func newChatModel(ctx context.Context, conf *config) (chatmodel.ToolCallingChatM return cm, nil } -func loadModelInfo(ctx context.Context, modelID int64) (*crossmodelmgr.Model, error) { +func loadModelInfo(ctx context.Context, manager modelmgr.Manager, modelID int64) (*modelmgr.Model, error) { if modelID == 0 { return nil, fmt.Errorf("modelID is required") } - models, err := crossmodelmgr.DefaultSVC().MGetModelByID(ctx, &modelmgr.MGetModelRequest{ + models, err := manager.MGetModelByID(ctx, &modelmgr.MGetModelRequest{ IDs: []int64{modelID}, }) diff --git a/backend/domain/agent/singleagent/service/single_agent_impl.go b/backend/domain/agent/singleagent/service/single_agent_impl.go index 7fdc714c..29b6cfcb 100644 --- a/backend/domain/agent/singleagent/service/single_agent_impl.go +++ b/backend/domain/agent/singleagent/service/single_agent_impl.go @@ -22,6 +22,7 @@ import ( "math/rand" "github.com/cloudwego/eino/compose" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" "github.com/jinzhu/copier" "github.com/cloudwego/eino/schema" @@ -44,6 +45,7 @@ type singleAgentImpl struct { } type Components struct { + ModelMgr modelmgr.Manager ModelFactory chatmodel.Factory AgentDraftRepo repository.SingleAgentDraftRepo @@ -106,6 +108,7 @@ func (s *singleAgentImpl) StreamExecute(ctx context.Context, req *entity.Execute Agent: ae, UserID: req.UserID, Identity: req.Identity, + ModelMgr: s.ModelMgr, ModelFactory: s.ModelFactory, CPStore: s.CPStore, } diff --git a/backend/domain/modelmgr/entity/chat_model.go b/backend/domain/modelmgr/entity/chat_model.go deleted file mode 100644 index 17b2e110..00000000 --- a/backend/domain/modelmgr/entity/chat_model.go +++ /dev/null @@ -1,41 +0,0 @@ -/* - * Copyright 2025 coze-dev Authors - * - * Licensed under the Apache License, Version 2.0 (the "License"); - * you may not use this file except in compliance with the License. - * You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package entity - -import "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" - -type Model struct { - *modelmgr.Model -} - -type ModelMeta = modelmgr.ModelMeta - -type ModelMetaStatus = modelmgr.ModelMetaStatus - -func (m *Model) FindParameter(name modelmgr.ParameterName) (*modelmgr.Parameter, bool) { - if len(m.DefaultParameters) == 0 { - return nil, false - } - - for _, param := range m.DefaultParameters { - if param.Name == name { - return param, true - } - } - - return nil, false -} diff --git a/backend/domain/modelmgr/entity/chat_model_test.go b/backend/domain/modelmgr/entity/chat_model_test.go deleted file mode 100644 index ccbaad2b..00000000 --- a/backend/domain/modelmgr/entity/chat_model_test.go +++ /dev/null @@ -1,89 +0,0 @@ -/* - * Copyright 2025 coze-dev Authors - * - * Licensed under the Apache License, Version 2.0 (the "License"); - * you may not use this file except in compliance with the License. - * You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package entity - -import ( - "encoding/json" - "testing" - - "github.com/stretchr/testify/assert" - - "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" -) - -func TestDefaultParameter(t *testing.T) { - dps := []*modelmgr.Parameter{ - { - Name: "temperature", - Label: &modelmgr.MultilingualText{ - ZH: "生成随机性", - EN: "Temperature", - }, - Desc: &modelmgr.MultilingualText{ - ZH: "- **temperature**: 调高温度会使得模型的输出更多样性和创新性，反之，降低温度会使输出内容更加遵循指令要求但减少多样性。建议不要与“Top p”同时调整。", - EN: "**Temperature**:\\n\\n- When you increase this value, the model outputs more diverse and innovative content; when you decrease it, the model outputs less diverse content that strictly follows the given instructions.\\n- It is recommended not to adjust this value with \\\"Top p\\\" at the same time.", - }, - Type: modelmgr.ValueTypeFloat, - Min: "0", - Max: "1", - Precision: 1, - DefaultVal: modelmgr.DefaultValue{ - modelmgr.DefaultTypeDefault: "1.0", - modelmgr.DefaultTypeCreative: "1", - modelmgr.DefaultTypeBalance: "0.8", - modelmgr.DefaultTypePrecise: "0.3", - }, - Style: modelmgr.DisplayStyle{ - Widget: modelmgr.WidgetSlider, - Label: &modelmgr.MultilingualText{ - ZH: "生成多样性", - EN: "Generation diversity", - }, - }, - }, - { - Name: "max_tokens", - Label: &modelmgr.MultilingualText{ - ZH: "最大回复长度", - EN: "Response max length", - }, - Desc: &modelmgr.MultilingualText{ - ZH: "控制模型输出的Tokens 长度上限。通常 100 Tokens 约等于 150 个中文汉字。", - EN: "You can specify the maximum length of the tokens output through this value. Typically, 100 tokens are approximately equal to 150 Chinese characters.", - }, - Type: modelmgr.ValueTypeInt, - Min: "1", - Max: "12288", - Precision: 0, - DefaultVal: modelmgr.DefaultValue{ - modelmgr.DefaultTypeDefault: "4096", - }, - Style: modelmgr.DisplayStyle{ - Widget: modelmgr.WidgetSlider, - Label: &modelmgr.MultilingualText{ - ZH: "输入及输出设置", - EN: "Input and output settings", - }, - }, - }, - } - - data, err := json.Marshal(dps) - assert.NoError(t, err) - - t.Logf("default parameters: %s", string(data)) -} diff --git a/backend/domain/modelmgr/entity/model.schema.go b/backend/domain/modelmgr/entity/model.schema.go deleted file mode 100644 index f19333ab..00000000 --- a/backend/domain/modelmgr/entity/model.schema.go +++ /dev/null @@ -1,23 +0,0 @@ -/* - * Copyright 2025 coze-dev Authors - * - * Licensed under the Apache License, Version 2.0 (the "License"); - * you may not use this file except in compliance with the License. - * You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package entity - -import ( - "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" -) - -type Capability = modelmgr.Capability diff --git a/backend/domain/modelmgr/interface.go b/backend/domain/modelmgr/interface.go deleted file mode 100644 index 543584f9..00000000 --- a/backend/domain/modelmgr/interface.go +++ /dev/null @@ -1,69 +0,0 @@ -/* - * Copyright 2025 coze-dev Authors - * - * Licensed under the Apache License, Version 2.0 (the "License"); - * you may not use this file except in compliance with the License. - * You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package modelmgr - -import ( - "context" - - "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr/entity" -) - -type Manager interface { - CreateModelMeta(ctx context.Context, meta *entity.ModelMeta) (*entity.ModelMeta, error) - UpdateModelMetaStatus(ctx context.Context, id int64, status entity.ModelMetaStatus) error - DeleteModelMeta(ctx context.Context, id int64) error - ListModelMeta(ctx context.Context, req *ListModelMetaRequest) (*ListModelMetaResponse, error) - MGetModelMetaByID(ctx context.Context, req *MGetModelMetaRequest) ([]*entity.ModelMeta, error) - - CreateModel(ctx context.Context, model *entity.Model) (*entity.Model, error) - DeleteModel(ctx context.Context, id int64) error - ListModel(ctx context.Context, req *ListModelRequest) (*ListModelResponse, error) - MGetModelByID(ctx context.Context, req *MGetModelRequest) ([]*entity.Model, error) -} - -type ListModelMetaRequest struct { - FuzzyModelName *string - Status []entity.ModelMetaStatus - Limit int - Cursor *string -} - -type ListModelMetaResponse struct { - ModelMetaList []*entity.ModelMeta - HasMore bool - NextCursor *string -} - -type MGetModelMetaRequest struct { - IDs []int64 -} - -type ListModelRequest struct { - FuzzyModelName *string - Status []modelmgr.ModelEntityStatus // default is default and in_use status - Limit int - Cursor *string -} - -type ListModelResponse struct { - ModelList []*entity.Model - HasMore bool - NextCursor *string -} - -type MGetModelRequest = modelmgr.MGetModelRequest diff --git a/backend/domain/modelmgr/internal/dal/dao/chat_model_entity.go b/backend/domain/modelmgr/internal/dal/dao/chat_model_entity.go deleted file mode 100644 index 50816f76..00000000 --- a/backend/domain/modelmgr/internal/dal/dao/chat_model_entity.go +++ /dev/null @@ -1,136 +0,0 @@ -/* - * Copyright 2025 coze-dev Authors - * - * Licensed under the Apache License, Version 2.0 (the "License"); - * you may not use this file except in compliance with the License. - * You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package dao - -import ( - "context" - "database/sql/driver" - "strconv" - - "gorm.io/gorm" - - "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr/internal/dal/model" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr/internal/dal/query" - "github.com/coze-dev/coze-studio/backend/pkg/lang/slices" - "github.com/coze-dev/coze-studio/backend/pkg/lang/sqlutil" -) - -type ModelEntityRepo interface { - Create(ctx context.Context, modelEntity *model.ModelEntity) error - Delete(ctx context.Context, id int64) error - List(ctx context.Context, fuzzyModelName *string, scenario *int64, status []modelmgr.ModelEntityStatus, - limit int, cursor *string) (resp []*model.ModelEntity, nextCursor *string, hasMore bool, err error) - MGet(ctx context.Context, ids []int64) ([]*model.ModelEntity, error) -} - -func NewModelEntityDAO(db *gorm.DB) ModelEntityRepo { - return &ModelEntityDAO{ - db: db, - query: query.Use(db), - } -} - -type ModelEntityDAO struct { - db *gorm.DB - query *query.Query -} - -func (m *ModelEntityDAO) Create(ctx context.Context, modelEntity *model.ModelEntity) error { - return m.query.ModelEntity.WithContext(ctx).Create(modelEntity) -} - -func (m *ModelEntityDAO) Delete(ctx context.Context, id int64) error { - me := m.query.ModelEntity - _, err := me.WithContext(ctx). - Debug(). - Where(me.ID.Eq(id)). - Delete() - - return err -} - -func (m *ModelEntityDAO) List(ctx context.Context, fuzzyModelName *string, scenario *int64, status []modelmgr.ModelEntityStatus, - limit int, cursor *string, -) (resp []*model.ModelEntity, nextCursor *string, hasMore bool, err error) { - me := m.query.ModelEntity - do := me.WithContext(ctx) - - if fuzzyModelName != nil { - do = do.Where(me.Name.Like(*fuzzyModelName)) - } - if scenario != nil { - do = do.Where(me.Scenario.Eq(sqlutil.DriverValue(*scenario))) - } - if len(status) > 0 { - vals := slices.Transform(status, func(a modelmgr.ModelEntityStatus) driver.Valuer { - return sqlutil.DriverValue(int64(a)) - }) - - do = do.Where(me.Status.In(vals...)) - } - if cursor != nil { - var id int64 - id, err = m.fromCursor(*cursor) - if err != nil { - return nil, nil, false, err - } - do = do.Where(me.ID.Lt(id)) - } - if limit == 0 { - limit = defaultLimit - } - - pos, err := do.Limit(limit).Order(me.ID.Desc()).Find() - if err != nil { - return nil, nil, false, err - } - - if len(pos) == 0 { - return nil, nil, false, nil - } - - hasMore = len(pos) == limit - if len(pos) > 0 { - nextCursor = m.toIDCursor(pos[len(pos)-1].ID) - } - - return pos, nextCursor, hasMore, nil -} - -func (m *ModelEntityDAO) MGet(ctx context.Context, ids []int64) ([]*model.ModelEntity, error) { - if len(ids) == 0 { - return nil, nil - } - - me := m.query.ModelEntity - pos, err := me.WithContext(ctx).Where(me.ID.In(ids...)).Find() - if err != nil { - return nil, err - } - - return pos, nil -} - -func (m *ModelEntityDAO) fromCursor(cursor string) (id int64, err error) { - return strconv.ParseInt(cursor, 10, 64) -} - -func (m *ModelEntityDAO) toIDCursor(id int64) (cursor *string) { - s := strconv.FormatInt(id, 10) - return &s -} diff --git a/backend/domain/modelmgr/internal/dal/dao/chat_model_meta.go b/backend/domain/modelmgr/internal/dal/dao/chat_model_meta.go deleted file mode 100644 index 4a1bf1d8..00000000 --- a/backend/domain/modelmgr/internal/dal/dao/chat_model_meta.go +++ /dev/null @@ -1,189 +0,0 @@ -/* - * Copyright 2025 coze-dev Authors - * - * Licensed under the Apache License, Version 2.0 (the "License"); - * you may not use this file except in compliance with the License. - * You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package dao - -import ( - "context" - "database/sql/driver" - "errors" - "strconv" - "time" - - "gorm.io/gorm" - - "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr/internal/dal/model" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr/internal/dal/query" - "github.com/coze-dev/coze-studio/backend/pkg/lang/slices" - "github.com/coze-dev/coze-studio/backend/pkg/lang/sqlutil" -) - -const ( - defaultLimit = 100 -) - -type ModelMetaRepo interface { - Create(ctx context.Context, meta *model.ModelMeta) error - UpdateStatus(ctx context.Context, id int64, status modelmgr.ModelMetaStatus) error - Delete(ctx context.Context, id int64) error - List(ctx context.Context, fuzzyShowName *string, status []modelmgr.ModelMetaStatus, limit int, cursor *string) ( - resp []*model.ModelMeta, nextCursor *string, hasMore bool, err error) - GetByID(ctx context.Context, id int64) (*model.ModelMeta, error) - MGetByID(ctx context.Context, ids []int64) ([]*model.ModelMeta, error) -} - -func NewModelMetaDAO(db *gorm.DB) ModelMetaRepo { - return &ModelMetaDAO{ - db: db, - query: query.Use(db), - } -} - -type ModelMetaDAO struct { - db *gorm.DB - query *query.Query -} - -func (m *ModelMetaDAO) Create(ctx context.Context, meta *model.ModelMeta) error { - return m.query.ModelMeta.WithContext(ctx).Create(meta) -} - -func (m *ModelMetaDAO) UpdateStatus(ctx context.Context, id int64, status modelmgr.ModelMetaStatus) error { - mm := m.query.ModelMeta - _, err := mm.WithContext(ctx). - Debug(). - Where(mm.ID.Eq(id)). - Select(mm.Status, mm.UpdatedAt). - Updates(&model.ModelMeta{ - Status: status, - UpdatedAt: time.Now().UnixMilli(), - }) - - return err -} - -func (m *ModelMetaDAO) Delete(ctx context.Context, id int64) error { - mm := m.query.ModelMeta - _, err := mm.WithContext(ctx). - Debug(). - Where(mm.ID.Eq(id)). - Delete() - - return err -} - -func (m *ModelMetaDAO) List(ctx context.Context, fuzzyShowName *string, status []modelmgr.ModelMetaStatus, limit int, cursor *string) ( - resp []*model.ModelMeta, nextCursor *string, hasMore bool, err error, -) { - mm := m.query.ModelMeta - do := mm.WithContext(ctx) - - if fuzzyShowName != nil { - do.Where(mm.ModelName.Like(*fuzzyShowName)) - } - - if len(status) > 0 { - vals := slices.Transform(status, func(a modelmgr.ModelMetaStatus) driver.Valuer { - return sqlutil.DriverValue(a) - }) - do.Where(mm.Status.In(vals...)) - } - - if cursor != nil { - id, err := m.fromCursor(*cursor) - if err != nil { - return nil, nil, false, err - } - - do.Where(mm.ID.Lt(id)) - } - - if limit == 0 { - limit = defaultLimit - } - - pos, err := do.Limit(limit).Order(mm.ID.Desc()).Find() - if err != nil { - return nil, nil, false, err - } - if len(pos) == 0 { - return nil, nil, false, nil - } - - hasMore = len(pos) == limit - if len(pos) > 0 { - nextCursor = m.toIDCursor(pos[len(pos)-1].ID) - } - - return pos, nextCursor, hasMore, nil -} - -func (m *ModelMetaDAO) GetByID(ctx context.Context, id int64) (*model.ModelMeta, error) { - mm := m.query.ModelMeta - po, err := mm.WithContext(ctx).Where(mm.ID.Eq(id)).Take() - if err != nil { - if errors.Is(err, gorm.ErrRecordNotFound) { - return nil, nil - } - - return nil, err - } - - return po, nil -} - -func (m *ModelMetaDAO) MGetByID(ctx context.Context, ids []int64) ([]*model.ModelMeta, error) { - if len(ids) == 0 { - return nil, nil - } - - mm := m.query.ModelMeta - do := mm.WithContext(ctx) - - pos, err := do.Where(mm.ID.In(ids...)).Find() - if err != nil { - return nil, err - } - - // todo::本意是按照ids的顺序返回结果，但是这里的查询结果是无序的，所以这里先不做排序 @zhaonan - // id2Idx := make(map[int64]int, len(ids)) - // for idx, id := range ids { - // id2Idx[id] = idx - // } - // - // resp := make([]*model.ModelMeta, 0, len(ids)) - // for _, po := range pos { - // idx, found := id2Idx[po.ID] - // if !found { // unexpected - // return nil, fmt.Errorf("[MGetByID] unexpected data found, id=%v", po.ID) - // } - // - // item := po - // resp[idx] = item - // } - - return pos, nil -} - -func (m *ModelMetaDAO) fromCursor(cursor string) (id int64, err error) { - return strconv.ParseInt(cursor, 10, 64) -} - -func (m *ModelMetaDAO) toIDCursor(id int64) (cursor *string) { - s := strconv.FormatInt(id, 10) - return &s -} diff --git a/backend/domain/modelmgr/internal/dal/model/model_entity.gen.go b/backend/domain/modelmgr/internal/dal/model/model_entity.gen.go deleted file mode 100644 index 997ea85c..00000000 --- a/backend/domain/modelmgr/internal/dal/model/model_entity.gen.go +++ /dev/null @@ -1,31 +0,0 @@ -// Code generated by gorm.io/gen. DO NOT EDIT. -// Code generated by gorm.io/gen. DO NOT EDIT. -// Code generated by gorm.io/gen. DO NOT EDIT. - -package model - -import ( - "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" - "gorm.io/gorm" -) - -const TableNameModelEntity = "model_entity" - -// ModelEntity 模型信息 -type ModelEntity struct { - ID int64 `gorm:"column:id;primaryKey;comment:主键ID" json:"id"` // 主键ID - MetaID int64 `gorm:"column:meta_id;not null;comment:模型元信息 id" json:"meta_id"` // 模型元信息 id - Name string `gorm:"column:name;not null;comment:名称" json:"name"` // 名称 - Description string `gorm:"column:description;comment:描述" json:"description"` // 描述 - DefaultParams []*modelmgr.Parameter `gorm:"column:default_params;comment:默认参数;serializer:json" json:"default_params"` // 默认参数 - Scenario modelmgr.Scenario `gorm:"column:scenario;not null;comment:模型应用场景;serializer:json" json:"scenario"` // 模型应用场景 - Status modelmgr.ModelEntityStatus `gorm:"column:status;not null;default:1;comment:模型状态;serializer:json" json:"status"` // 模型状态 - CreatedAt int64 `gorm:"column:created_at;not null;autoCreateTime:milli;comment:Create Time in Milliseconds" json:"created_at"` // Create Time in Milliseconds - UpdatedAt int64 `gorm:"column:updated_at;not null;autoUpdateTime:milli;comment:Update Time in Milliseconds" json:"updated_at"` // Update Time in Milliseconds - DeletedAt gorm.DeletedAt `gorm:"column:deleted_at;comment:Delete Time in Milliseconds" json:"deleted_at"` // Delete Time in Milliseconds -} - -// TableName ModelEntity's table name -func (*ModelEntity) TableName() string { - return TableNameModelEntity -} diff --git a/backend/domain/modelmgr/internal/dal/model/model_meta.gen.go b/backend/domain/modelmgr/internal/dal/model/model_meta.gen.go deleted file mode 100644 index f008dfb6..00000000 --- a/backend/domain/modelmgr/internal/dal/model/model_meta.gen.go +++ /dev/null @@ -1,34 +0,0 @@ -// Code generated by gorm.io/gen. DO NOT EDIT. -// Code generated by gorm.io/gen. DO NOT EDIT. -// Code generated by gorm.io/gen. DO NOT EDIT. - -package model - -import ( - "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" - "github.com/coze-dev/coze-studio/backend/infra/contract/chatmodel" - "gorm.io/gorm" -) - -const TableNameModelMeta = "model_meta" - -// ModelMeta 模型元信息 -type ModelMeta struct { - ID int64 `gorm:"column:id;primaryKey;comment:主键ID" json:"id"` // 主键ID - ModelName string `gorm:"column:model_name;not null;comment:模型名称" json:"model_name"` // 模型名称 - Protocol string `gorm:"column:protocol;not null;comment:模型协议" json:"protocol"` // 模型协议 - IconURI string `gorm:"column:icon_uri;not null;comment:Icon URI" json:"icon_uri"` // Icon URI - Capability *modelmgr.Capability `gorm:"column:capability;comment:模型能力;serializer:json" json:"capability"` // 模型能力 - ConnConfig *chatmodel.Config `gorm:"column:conn_config;comment:模型连接配置;serializer:json" json:"conn_config"` // 模型连接配置 - Status modelmgr.ModelMetaStatus `gorm:"column:status;not null;default:1;comment:模型状态;serializer:json" json:"status"` // 模型状态 - Description string `gorm:"column:description;not null;comment:模型描述" json:"description"` // 模型描述 - CreatedAt int64 `gorm:"column:created_at;not null;autoCreateTime:milli;comment:Create Time in Milliseconds" json:"created_at"` // Create Time in Milliseconds - UpdatedAt int64 `gorm:"column:updated_at;not null;autoUpdateTime:milli;comment:Update Time in Milliseconds" json:"updated_at"` // Update Time in Milliseconds - DeletedAt gorm.DeletedAt `gorm:"column:deleted_at;comment:Delete Time in Milliseconds" json:"deleted_at"` // Delete Time in Milliseconds - IconURL string `gorm:"column:icon_url;not null;comment:Icon URL" json:"icon_url"` // Icon URL -} - -// TableName ModelMeta's table name -func (*ModelMeta) TableName() string { - return TableNameModelMeta -} diff --git a/backend/domain/modelmgr/internal/dal/query/gen.go b/backend/domain/modelmgr/internal/dal/query/gen.go deleted file mode 100644 index f01a1909..00000000 --- a/backend/domain/modelmgr/internal/dal/query/gen.go +++ /dev/null @@ -1,111 +0,0 @@ -// Code generated by gorm.io/gen. DO NOT EDIT. -// Code generated by gorm.io/gen. DO NOT EDIT. -// Code generated by gorm.io/gen. DO NOT EDIT. - -package query - -import ( - "context" - "database/sql" - - "gorm.io/gorm" - - "gorm.io/gen" - - "gorm.io/plugin/dbresolver" -) - -var ( - Q = new(Query) - ModelEntity *modelEntity - ModelMeta *modelMeta -) - -func SetDefault(db *gorm.DB, opts ...gen.DOOption) { - *Q = *Use(db, opts...) - ModelEntity = &Q.ModelEntity - ModelMeta = &Q.ModelMeta -} - -func Use(db *gorm.DB, opts ...gen.DOOption) *Query { - return &Query{ - db: db, - ModelEntity: newModelEntity(db, opts...), - ModelMeta: newModelMeta(db, opts...), - } -} - -type Query struct { - db *gorm.DB - - ModelEntity modelEntity - ModelMeta modelMeta -} - -func (q *Query) Available() bool { return q.db != nil } - -func (q *Query) clone(db *gorm.DB) *Query { - return &Query{ - db: db, - ModelEntity: q.ModelEntity.clone(db), - ModelMeta: q.ModelMeta.clone(db), - } -} - -func (q *Query) ReadDB() *Query { - return q.ReplaceDB(q.db.Clauses(dbresolver.Read)) -} - -func (q *Query) WriteDB() *Query { - return q.ReplaceDB(q.db.Clauses(dbresolver.Write)) -} - -func (q *Query) ReplaceDB(db *gorm.DB) *Query { - return &Query{ - db: db, - ModelEntity: q.ModelEntity.replaceDB(db), - ModelMeta: q.ModelMeta.replaceDB(db), - } -} - -type queryCtx struct { - ModelEntity IModelEntityDo - ModelMeta IModelMetaDo -} - -func (q *Query) WithContext(ctx context.Context) *queryCtx { - return &queryCtx{ - ModelEntity: q.ModelEntity.WithContext(ctx), - ModelMeta: q.ModelMeta.WithContext(ctx), - } -} - -func (q *Query) Transaction(fc func(tx *Query) error, opts ...*sql.TxOptions) error { - return q.db.Transaction(func(tx *gorm.DB) error { return fc(q.clone(tx)) }, opts...) -} - -func (q *Query) Begin(opts ...*sql.TxOptions) *QueryTx { - tx := q.db.Begin(opts...) - return &QueryTx{Query: q.clone(tx), Error: tx.Error} -} - -type QueryTx struct { - *Query - Error error -} - -func (q *QueryTx) Commit() error { - return q.db.Commit().Error -} - -func (q *QueryTx) Rollback() error { - return q.db.Rollback().Error -} - -func (q *QueryTx) SavePoint(name string) error { - return q.db.SavePoint(name).Error -} - -func (q *QueryTx) RollbackTo(name string) error { - return q.db.RollbackTo(name).Error -} diff --git a/backend/domain/modelmgr/internal/dal/query/model_entity.gen.go b/backend/domain/modelmgr/internal/dal/query/model_entity.gen.go deleted file mode 100644 index ac8e6c58..00000000 --- a/backend/domain/modelmgr/internal/dal/query/model_entity.gen.go +++ /dev/null @@ -1,417 +0,0 @@ -// Code generated by gorm.io/gen. DO NOT EDIT. -// Code generated by gorm.io/gen. DO NOT EDIT. -// Code generated by gorm.io/gen. DO NOT EDIT. - -package query - -import ( - "context" - - "gorm.io/gorm" - "gorm.io/gorm/clause" - "gorm.io/gorm/schema" - - "gorm.io/gen" - "gorm.io/gen/field" - - "gorm.io/plugin/dbresolver" - - "github.com/coze-dev/coze-studio/backend/domain/modelmgr/internal/dal/model" -) - -func newModelEntity(db *gorm.DB, opts ...gen.DOOption) modelEntity { - _modelEntity := modelEntity{} - - _modelEntity.modelEntityDo.UseDB(db, opts...) - _modelEntity.modelEntityDo.UseModel(&model.ModelEntity{}) - - tableName := _modelEntity.modelEntityDo.TableName() - _modelEntity.ALL = field.NewAsterisk(tableName) - _modelEntity.ID = field.NewInt64(tableName, "id") - _modelEntity.MetaID = field.NewInt64(tableName, "meta_id") - _modelEntity.Name = field.NewString(tableName, "name") - _modelEntity.Description = field.NewString(tableName, "description") - _modelEntity.DefaultParams = field.NewField(tableName, "default_params") - _modelEntity.Scenario = field.NewField(tableName, "scenario") - _modelEntity.Status = field.NewField(tableName, "status") - _modelEntity.CreatedAt = field.NewInt64(tableName, "created_at") - _modelEntity.UpdatedAt = field.NewInt64(tableName, "updated_at") - _modelEntity.DeletedAt = field.NewField(tableName, "deleted_at") - - _modelEntity.fillFieldMap() - - return _modelEntity -} - -// modelEntity 模型信息 -type modelEntity struct { - modelEntityDo - - ALL field.Asterisk - ID field.Int64 // 主键ID - MetaID field.Int64 // 模型元信息 id - Name field.String // 名称 - Description field.String // 描述 - DefaultParams field.Field // 默认参数 - Scenario field.Field // 模型应用场景 - Status field.Field // 模型状态 - CreatedAt field.Int64 // Create Time in Milliseconds - UpdatedAt field.Int64 // Update Time in Milliseconds - DeletedAt field.Field // Delete Time in Milliseconds - - fieldMap map[string]field.Expr -} - -func (m modelEntity) Table(newTableName string) *modelEntity { - m.modelEntityDo.UseTable(newTableName) - return m.updateTableName(newTableName) -} - -func (m modelEntity) As(alias string) *modelEntity { - m.modelEntityDo.DO = *(m.modelEntityDo.As(alias).(*gen.DO)) - return m.updateTableName(alias) -} - -func (m *modelEntity) updateTableName(table string) *modelEntity { - m.ALL = field.NewAsterisk(table) - m.ID = field.NewInt64(table, "id") - m.MetaID = field.NewInt64(table, "meta_id") - m.Name = field.NewString(table, "name") - m.Description = field.NewString(table, "description") - m.DefaultParams = field.NewField(table, "default_params") - m.Scenario = field.NewField(table, "scenario") - m.Status = field.NewField(table, "status") - m.CreatedAt = field.NewInt64(table, "created_at") - m.UpdatedAt = field.NewInt64(table, "updated_at") - m.DeletedAt = field.NewField(table, "deleted_at") - - m.fillFieldMap() - - return m -} - -func (m *modelEntity) GetFieldByName(fieldName string) (field.OrderExpr, bool) { - _f, ok := m.fieldMap[fieldName] - if !ok || _f == nil { - return nil, false - } - _oe, ok := _f.(field.OrderExpr) - return _oe, ok -} - -func (m *modelEntity) fillFieldMap() { - m.fieldMap = make(map[string]field.Expr, 10) - m.fieldMap["id"] = m.ID - m.fieldMap["meta_id"] = m.MetaID - m.fieldMap["name"] = m.Name - m.fieldMap["description"] = m.Description - m.fieldMap["default_params"] = m.DefaultParams - m.fieldMap["scenario"] = m.Scenario - m.fieldMap["status"] = m.Status - m.fieldMap["created_at"] = m.CreatedAt - m.fieldMap["updated_at"] = m.UpdatedAt - m.fieldMap["deleted_at"] = m.DeletedAt -} - -func (m modelEntity) clone(db *gorm.DB) modelEntity { - m.modelEntityDo.ReplaceConnPool(db.Statement.ConnPool) - return m -} - -func (m modelEntity) replaceDB(db *gorm.DB) modelEntity { - m.modelEntityDo.ReplaceDB(db) - return m -} - -type modelEntityDo struct{ gen.DO } - -type IModelEntityDo interface { - gen.SubQuery - Debug() IModelEntityDo - WithContext(ctx context.Context) IModelEntityDo - WithResult(fc func(tx gen.Dao)) gen.ResultInfo - ReplaceDB(db *gorm.DB) - ReadDB() IModelEntityDo - WriteDB() IModelEntityDo - As(alias string) gen.Dao - Session(config *gorm.Session) IModelEntityDo - Columns(cols ...field.Expr) gen.Columns - Clauses(conds ...clause.Expression) IModelEntityDo - Not(conds ...gen.Condition) IModelEntityDo - Or(conds ...gen.Condition) IModelEntityDo - Select(conds ...field.Expr) IModelEntityDo - Where(conds ...gen.Condition) IModelEntityDo - Order(conds ...field.Expr) IModelEntityDo - Distinct(cols ...field.Expr) IModelEntityDo - Omit(cols ...field.Expr) IModelEntityDo - Join(table schema.Tabler, on ...field.Expr) IModelEntityDo - LeftJoin(table schema.Tabler, on ...field.Expr) IModelEntityDo - RightJoin(table schema.Tabler, on ...field.Expr) IModelEntityDo - Group(cols ...field.Expr) IModelEntityDo - Having(conds ...gen.Condition) IModelEntityDo - Limit(limit int) IModelEntityDo - Offset(offset int) IModelEntityDo - Count() (count int64, err error) - Scopes(funcs ...func(gen.Dao) gen.Dao) IModelEntityDo - Unscoped() IModelEntityDo - Create(values ...*model.ModelEntity) error - CreateInBatches(values []*model.ModelEntity, batchSize int) error - Save(values ...*model.ModelEntity) error - First() (*model.ModelEntity, error) - Take() (*model.ModelEntity, error) - Last() (*model.ModelEntity, error) - Find() ([]*model.ModelEntity, error) - FindInBatch(batchSize int, fc func(tx gen.Dao, batch int) error) (results []*model.ModelEntity, err error) - FindInBatches(result *[]*model.ModelEntity, batchSize int, fc func(tx gen.Dao, batch int) error) error - Pluck(column field.Expr, dest interface{}) error - Delete(...*model.ModelEntity) (info gen.ResultInfo, err error) - Update(column field.Expr, value interface{}) (info gen.ResultInfo, err error) - UpdateSimple(columns ...field.AssignExpr) (info gen.ResultInfo, err error) - Updates(value interface{}) (info gen.ResultInfo, err error) - UpdateColumn(column field.Expr, value interface{}) (info gen.ResultInfo, err error) - UpdateColumnSimple(columns ...field.AssignExpr) (info gen.ResultInfo, err error) - UpdateColumns(value interface{}) (info gen.ResultInfo, err error) - UpdateFrom(q gen.SubQuery) gen.Dao - Attrs(attrs ...field.AssignExpr) IModelEntityDo - Assign(attrs ...field.AssignExpr) IModelEntityDo - Joins(fields ...field.RelationField) IModelEntityDo - Preload(fields ...field.RelationField) IModelEntityDo - FirstOrInit() (*model.ModelEntity, error) - FirstOrCreate() (*model.ModelEntity, error) - FindByPage(offset int, limit int) (result []*model.ModelEntity, count int64, err error) - ScanByPage(result interface{}, offset int, limit int) (count int64, err error) - Scan(result interface{}) (err error) - Returning(value interface{}, columns ...string) IModelEntityDo - UnderlyingDB() *gorm.DB - schema.Tabler -} - -func (m modelEntityDo) Debug() IModelEntityDo { - return m.withDO(m.DO.Debug()) -} - -func (m modelEntityDo) WithContext(ctx context.Context) IModelEntityDo { - return m.withDO(m.DO.WithContext(ctx)) -} - -func (m modelEntityDo) ReadDB() IModelEntityDo { - return m.Clauses(dbresolver.Read) -} - -func (m modelEntityDo) WriteDB() IModelEntityDo { - return m.Clauses(dbresolver.Write) -} - -func (m modelEntityDo) Session(config *gorm.Session) IModelEntityDo { - return m.withDO(m.DO.Session(config)) -} - -func (m modelEntityDo) Clauses(conds ...clause.Expression) IModelEntityDo { - return m.withDO(m.DO.Clauses(conds...)) -} - -func (m modelEntityDo) Returning(value interface{}, columns ...string) IModelEntityDo { - return m.withDO(m.DO.Returning(value, columns...)) -} - -func (m modelEntityDo) Not(conds ...gen.Condition) IModelEntityDo { - return m.withDO(m.DO.Not(conds...)) -} - -func (m modelEntityDo) Or(conds ...gen.Condition) IModelEntityDo { - return m.withDO(m.DO.Or(conds...)) -} - -func (m modelEntityDo) Select(conds ...field.Expr) IModelEntityDo { - return m.withDO(m.DO.Select(conds...)) -} - -func (m modelEntityDo) Where(conds ...gen.Condition) IModelEntityDo { - return m.withDO(m.DO.Where(conds...)) -} - -func (m modelEntityDo) Order(conds ...field.Expr) IModelEntityDo { - return m.withDO(m.DO.Order(conds...)) -} - -func (m modelEntityDo) Distinct(cols ...field.Expr) IModelEntityDo { - return m.withDO(m.DO.Distinct(cols...)) -} - -func (m modelEntityDo) Omit(cols ...field.Expr) IModelEntityDo { - return m.withDO(m.DO.Omit(cols...)) -} - -func (m modelEntityDo) Join(table schema.Tabler, on ...field.Expr) IModelEntityDo { - return m.withDO(m.DO.Join(table, on...)) -} - -func (m modelEntityDo) LeftJoin(table schema.Tabler, on ...field.Expr) IModelEntityDo { - return m.withDO(m.DO.LeftJoin(table, on...)) -} - -func (m modelEntityDo) RightJoin(table schema.Tabler, on ...field.Expr) IModelEntityDo { - return m.withDO(m.DO.RightJoin(table, on...)) -} - -func (m modelEntityDo) Group(cols ...field.Expr) IModelEntityDo { - return m.withDO(m.DO.Group(cols...)) -} - -func (m modelEntityDo) Having(conds ...gen.Condition) IModelEntityDo { - return m.withDO(m.DO.Having(conds...)) -} - -func (m modelEntityDo) Limit(limit int) IModelEntityDo { - return m.withDO(m.DO.Limit(limit)) -} - -func (m modelEntityDo) Offset(offset int) IModelEntityDo { - return m.withDO(m.DO.Offset(offset)) -} - -func (m modelEntityDo) Scopes(funcs ...func(gen.Dao) gen.Dao) IModelEntityDo { - return m.withDO(m.DO.Scopes(funcs...)) -} - -func (m modelEntityDo) Unscoped() IModelEntityDo { - return m.withDO(m.DO.Unscoped()) -} - -func (m modelEntityDo) Create(values ...*model.ModelEntity) error { - if len(values) == 0 { - return nil - } - return m.DO.Create(values) -} - -func (m modelEntityDo) CreateInBatches(values []*model.ModelEntity, batchSize int) error { - return m.DO.CreateInBatches(values, batchSize) -} - -// Save : !!! underlying implementation is different with GORM -// The method is equivalent to executing the statement: db.Clauses(clause.OnConflict{UpdateAll: true}).Create(values) -func (m modelEntityDo) Save(values ...*model.ModelEntity) error { - if len(values) == 0 { - return nil - } - return m.DO.Save(values) -} - -func (m modelEntityDo) First() (*model.ModelEntity, error) { - if result, err := m.DO.First(); err != nil { - return nil, err - } else { - return result.(*model.ModelEntity), nil - } -} - -func (m modelEntityDo) Take() (*model.ModelEntity, error) { - if result, err := m.DO.Take(); err != nil { - return nil, err - } else { - return result.(*model.ModelEntity), nil - } -} - -func (m modelEntityDo) Last() (*model.ModelEntity, error) { - if result, err := m.DO.Last(); err != nil { - return nil, err - } else { - return result.(*model.ModelEntity), nil - } -} - -func (m modelEntityDo) Find() ([]*model.ModelEntity, error) { - result, err := m.DO.Find() - return result.([]*model.ModelEntity), err -} - -func (m modelEntityDo) FindInBatch(batchSize int, fc func(tx gen.Dao, batch int) error) (results []*model.ModelEntity, err error) { - buf := make([]*model.ModelEntity, 0, batchSize) - err = m.DO.FindInBatches(&buf, batchSize, func(tx gen.Dao, batch int) error { - defer func() { results = append(results, buf...) }() - return fc(tx, batch) - }) - return results, err -} - -func (m modelEntityDo) FindInBatches(result *[]*model.ModelEntity, batchSize int, fc func(tx gen.Dao, batch int) error) error { - return m.DO.FindInBatches(result, batchSize, fc) -} - -func (m modelEntityDo) Attrs(attrs ...field.AssignExpr) IModelEntityDo { - return m.withDO(m.DO.Attrs(attrs...)) -} - -func (m modelEntityDo) Assign(attrs ...field.AssignExpr) IModelEntityDo { - return m.withDO(m.DO.Assign(attrs...)) -} - -func (m modelEntityDo) Joins(fields ...field.RelationField) IModelEntityDo { - for _, _f := range fields { - m = *m.withDO(m.DO.Joins(_f)) - } - return &m -} - -func (m modelEntityDo) Preload(fields ...field.RelationField) IModelEntityDo { - for _, _f := range fields { - m = *m.withDO(m.DO.Preload(_f)) - } - return &m -} - -func (m modelEntityDo) FirstOrInit() (*model.ModelEntity, error) { - if result, err := m.DO.FirstOrInit(); err != nil { - return nil, err - } else { - return result.(*model.ModelEntity), nil - } -} - -func (m modelEntityDo) FirstOrCreate() (*model.ModelEntity, error) { - if result, err := m.DO.FirstOrCreate(); err != nil { - return nil, err - } else { - return result.(*model.ModelEntity), nil - } -} - -func (m modelEntityDo) FindByPage(offset int, limit int) (result []*model.ModelEntity, count int64, err error) { - result, err = m.Offset(offset).Limit(limit).Find() - if err != nil { - return - } - - if size := len(result); 0 < limit && 0 < size && size < limit { - count = int64(size + offset) - return - } - - count, err = m.Offset(-1).Limit(-1).Count() - return -} - -func (m modelEntityDo) ScanByPage(result interface{}, offset int, limit int) (count int64, err error) { - count, err = m.Count() - if err != nil { - return - } - - err = m.Offset(offset).Limit(limit).Scan(result) - return -} - -func (m modelEntityDo) Scan(result interface{}) (err error) { - return m.DO.Scan(result) -} - -func (m modelEntityDo) Delete(models ...*model.ModelEntity) (result gen.ResultInfo, err error) { - return m.DO.Delete(models) -} - -func (m *modelEntityDo) withDO(do gen.Dao) *modelEntityDo { - m.DO = *do.(*gen.DO) - return m -} diff --git a/backend/domain/modelmgr/internal/dal/query/model_meta.gen.go b/backend/domain/modelmgr/internal/dal/query/model_meta.gen.go deleted file mode 100644 index 36c48e29..00000000 --- a/backend/domain/modelmgr/internal/dal/query/model_meta.gen.go +++ /dev/null @@ -1,425 +0,0 @@ -// Code generated by gorm.io/gen. DO NOT EDIT. -// Code generated by gorm.io/gen. DO NOT EDIT. -// Code generated by gorm.io/gen. DO NOT EDIT. - -package query - -import ( - "context" - - "gorm.io/gorm" - "gorm.io/gorm/clause" - "gorm.io/gorm/schema" - - "gorm.io/gen" - "gorm.io/gen/field" - - "gorm.io/plugin/dbresolver" - - "github.com/coze-dev/coze-studio/backend/domain/modelmgr/internal/dal/model" -) - -func newModelMeta(db *gorm.DB, opts ...gen.DOOption) modelMeta { - _modelMeta := modelMeta{} - - _modelMeta.modelMetaDo.UseDB(db, opts...) - _modelMeta.modelMetaDo.UseModel(&model.ModelMeta{}) - - tableName := _modelMeta.modelMetaDo.TableName() - _modelMeta.ALL = field.NewAsterisk(tableName) - _modelMeta.ID = field.NewInt64(tableName, "id") - _modelMeta.ModelName = field.NewString(tableName, "model_name") - _modelMeta.Protocol = field.NewString(tableName, "protocol") - _modelMeta.IconURI = field.NewString(tableName, "icon_uri") - _modelMeta.Capability = field.NewField(tableName, "capability") - _modelMeta.ConnConfig = field.NewField(tableName, "conn_config") - _modelMeta.Status = field.NewField(tableName, "status") - _modelMeta.Description = field.NewString(tableName, "description") - _modelMeta.CreatedAt = field.NewInt64(tableName, "created_at") - _modelMeta.UpdatedAt = field.NewInt64(tableName, "updated_at") - _modelMeta.DeletedAt = field.NewField(tableName, "deleted_at") - _modelMeta.IconURL = field.NewString(tableName, "icon_url") - - _modelMeta.fillFieldMap() - - return _modelMeta -} - -// modelMeta 模型元信息 -type modelMeta struct { - modelMetaDo - - ALL field.Asterisk - ID field.Int64 // 主键ID - ModelName field.String // 模型名称 - Protocol field.String // 模型协议 - IconURI field.String // Icon URI - Capability field.Field // 模型能力 - ConnConfig field.Field // 模型连接配置 - Status field.Field // 模型状态 - Description field.String // 模型描述 - CreatedAt field.Int64 // Create Time in Milliseconds - UpdatedAt field.Int64 // Update Time in Milliseconds - DeletedAt field.Field // Delete Time in Milliseconds - IconURL field.String // Icon URL - - fieldMap map[string]field.Expr -} - -func (m modelMeta) Table(newTableName string) *modelMeta { - m.modelMetaDo.UseTable(newTableName) - return m.updateTableName(newTableName) -} - -func (m modelMeta) As(alias string) *modelMeta { - m.modelMetaDo.DO = *(m.modelMetaDo.As(alias).(*gen.DO)) - return m.updateTableName(alias) -} - -func (m *modelMeta) updateTableName(table string) *modelMeta { - m.ALL = field.NewAsterisk(table) - m.ID = field.NewInt64(table, "id") - m.ModelName = field.NewString(table, "model_name") - m.Protocol = field.NewString(table, "protocol") - m.IconURI = field.NewString(table, "icon_uri") - m.Capability = field.NewField(table, "capability") - m.ConnConfig = field.NewField(table, "conn_config") - m.Status = field.NewField(table, "status") - m.Description = field.NewString(table, "description") - m.CreatedAt = field.NewInt64(table, "created_at") - m.UpdatedAt = field.NewInt64(table, "updated_at") - m.DeletedAt = field.NewField(table, "deleted_at") - m.IconURL = field.NewString(table, "icon_url") - - m.fillFieldMap() - - return m -} - -func (m *modelMeta) GetFieldByName(fieldName string) (field.OrderExpr, bool) { - _f, ok := m.fieldMap[fieldName] - if !ok || _f == nil { - return nil, false - } - _oe, ok := _f.(field.OrderExpr) - return _oe, ok -} - -func (m *modelMeta) fillFieldMap() { - m.fieldMap = make(map[string]field.Expr, 12) - m.fieldMap["id"] = m.ID - m.fieldMap["model_name"] = m.ModelName - m.fieldMap["protocol"] = m.Protocol - m.fieldMap["icon_uri"] = m.IconURI - m.fieldMap["capability"] = m.Capability - m.fieldMap["conn_config"] = m.ConnConfig - m.fieldMap["status"] = m.Status - m.fieldMap["description"] = m.Description - m.fieldMap["created_at"] = m.CreatedAt - m.fieldMap["updated_at"] = m.UpdatedAt - m.fieldMap["deleted_at"] = m.DeletedAt - m.fieldMap["icon_url"] = m.IconURL -} - -func (m modelMeta) clone(db *gorm.DB) modelMeta { - m.modelMetaDo.ReplaceConnPool(db.Statement.ConnPool) - return m -} - -func (m modelMeta) replaceDB(db *gorm.DB) modelMeta { - m.modelMetaDo.ReplaceDB(db) - return m -} - -type modelMetaDo struct{ gen.DO } - -type IModelMetaDo interface { - gen.SubQuery - Debug() IModelMetaDo - WithContext(ctx context.Context) IModelMetaDo - WithResult(fc func(tx gen.Dao)) gen.ResultInfo - ReplaceDB(db *gorm.DB) - ReadDB() IModelMetaDo - WriteDB() IModelMetaDo - As(alias string) gen.Dao - Session(config *gorm.Session) IModelMetaDo - Columns(cols ...field.Expr) gen.Columns - Clauses(conds ...clause.Expression) IModelMetaDo - Not(conds ...gen.Condition) IModelMetaDo - Or(conds ...gen.Condition) IModelMetaDo - Select(conds ...field.Expr) IModelMetaDo - Where(conds ...gen.Condition) IModelMetaDo - Order(conds ...field.Expr) IModelMetaDo - Distinct(cols ...field.Expr) IModelMetaDo - Omit(cols ...field.Expr) IModelMetaDo - Join(table schema.Tabler, on ...field.Expr) IModelMetaDo - LeftJoin(table schema.Tabler, on ...field.Expr) IModelMetaDo - RightJoin(table schema.Tabler, on ...field.Expr) IModelMetaDo - Group(cols ...field.Expr) IModelMetaDo - Having(conds ...gen.Condition) IModelMetaDo - Limit(limit int) IModelMetaDo - Offset(offset int) IModelMetaDo - Count() (count int64, err error) - Scopes(funcs ...func(gen.Dao) gen.Dao) IModelMetaDo - Unscoped() IModelMetaDo - Create(values ...*model.ModelMeta) error - CreateInBatches(values []*model.ModelMeta, batchSize int) error - Save(values ...*model.ModelMeta) error - First() (*model.ModelMeta, error) - Take() (*model.ModelMeta, error) - Last() (*model.ModelMeta, error) - Find() ([]*model.ModelMeta, error) - FindInBatch(batchSize int, fc func(tx gen.Dao, batch int) error) (results []*model.ModelMeta, err error) - FindInBatches(result *[]*model.ModelMeta, batchSize int, fc func(tx gen.Dao, batch int) error) error - Pluck(column field.Expr, dest interface{}) error - Delete(...*model.ModelMeta) (info gen.ResultInfo, err error) - Update(column field.Expr, value interface{}) (info gen.ResultInfo, err error) - UpdateSimple(columns ...field.AssignExpr) (info gen.ResultInfo, err error) - Updates(value interface{}) (info gen.ResultInfo, err error) - UpdateColumn(column field.Expr, value interface{}) (info gen.ResultInfo, err error) - UpdateColumnSimple(columns ...field.AssignExpr) (info gen.ResultInfo, err error) - UpdateColumns(value interface{}) (info gen.ResultInfo, err error) - UpdateFrom(q gen.SubQuery) gen.Dao - Attrs(attrs ...field.AssignExpr) IModelMetaDo - Assign(attrs ...field.AssignExpr) IModelMetaDo - Joins(fields ...field.RelationField) IModelMetaDo - Preload(fields ...field.RelationField) IModelMetaDo - FirstOrInit() (*model.ModelMeta, error) - FirstOrCreate() (*model.ModelMeta, error) - FindByPage(offset int, limit int) (result []*model.ModelMeta, count int64, err error) - ScanByPage(result interface{}, offset int, limit int) (count int64, err error) - Scan(result interface{}) (err error) - Returning(value interface{}, columns ...string) IModelMetaDo - UnderlyingDB() *gorm.DB - schema.Tabler -} - -func (m modelMetaDo) Debug() IModelMetaDo { - return m.withDO(m.DO.Debug()) -} - -func (m modelMetaDo) WithContext(ctx context.Context) IModelMetaDo { - return m.withDO(m.DO.WithContext(ctx)) -} - -func (m modelMetaDo) ReadDB() IModelMetaDo { - return m.Clauses(dbresolver.Read) -} - -func (m modelMetaDo) WriteDB() IModelMetaDo { - return m.Clauses(dbresolver.Write) -} - -func (m modelMetaDo) Session(config *gorm.Session) IModelMetaDo { - return m.withDO(m.DO.Session(config)) -} - -func (m modelMetaDo) Clauses(conds ...clause.Expression) IModelMetaDo { - return m.withDO(m.DO.Clauses(conds...)) -} - -func (m modelMetaDo) Returning(value interface{}, columns ...string) IModelMetaDo { - return m.withDO(m.DO.Returning(value, columns...)) -} - -func (m modelMetaDo) Not(conds ...gen.Condition) IModelMetaDo { - return m.withDO(m.DO.Not(conds...)) -} - -func (m modelMetaDo) Or(conds ...gen.Condition) IModelMetaDo { - return m.withDO(m.DO.Or(conds...)) -} - -func (m modelMetaDo) Select(conds ...field.Expr) IModelMetaDo { - return m.withDO(m.DO.Select(conds...)) -} - -func (m modelMetaDo) Where(conds ...gen.Condition) IModelMetaDo { - return m.withDO(m.DO.Where(conds...)) -} - -func (m modelMetaDo) Order(conds ...field.Expr) IModelMetaDo { - return m.withDO(m.DO.Order(conds...)) -} - -func (m modelMetaDo) Distinct(cols ...field.Expr) IModelMetaDo { - return m.withDO(m.DO.Distinct(cols...)) -} - -func (m modelMetaDo) Omit(cols ...field.Expr) IModelMetaDo { - return m.withDO(m.DO.Omit(cols...)) -} - -func (m modelMetaDo) Join(table schema.Tabler, on ...field.Expr) IModelMetaDo { - return m.withDO(m.DO.Join(table, on...)) -} - -func (m modelMetaDo) LeftJoin(table schema.Tabler, on ...field.Expr) IModelMetaDo { - return m.withDO(m.DO.LeftJoin(table, on...)) -} - -func (m modelMetaDo) RightJoin(table schema.Tabler, on ...field.Expr) IModelMetaDo { - return m.withDO(m.DO.RightJoin(table, on...)) -} - -func (m modelMetaDo) Group(cols ...field.Expr) IModelMetaDo { - return m.withDO(m.DO.Group(cols...)) -} - -func (m modelMetaDo) Having(conds ...gen.Condition) IModelMetaDo { - return m.withDO(m.DO.Having(conds...)) -} - -func (m modelMetaDo) Limit(limit int) IModelMetaDo { - return m.withDO(m.DO.Limit(limit)) -} - -func (m modelMetaDo) Offset(offset int) IModelMetaDo { - return m.withDO(m.DO.Offset(offset)) -} - -func (m modelMetaDo) Scopes(funcs ...func(gen.Dao) gen.Dao) IModelMetaDo { - return m.withDO(m.DO.Scopes(funcs...)) -} - -func (m modelMetaDo) Unscoped() IModelMetaDo { - return m.withDO(m.DO.Unscoped()) -} - -func (m modelMetaDo) Create(values ...*model.ModelMeta) error { - if len(values) == 0 { - return nil - } - return m.DO.Create(values) -} - -func (m modelMetaDo) CreateInBatches(values []*model.ModelMeta, batchSize int) error { - return m.DO.CreateInBatches(values, batchSize) -} - -// Save : !!! underlying implementation is different with GORM -// The method is equivalent to executing the statement: db.Clauses(clause.OnConflict{UpdateAll: true}).Create(values) -func (m modelMetaDo) Save(values ...*model.ModelMeta) error { - if len(values) == 0 { - return nil - } - return m.DO.Save(values) -} - -func (m modelMetaDo) First() (*model.ModelMeta, error) { - if result, err := m.DO.First(); err != nil { - return nil, err - } else { - return result.(*model.ModelMeta), nil - } -} - -func (m modelMetaDo) Take() (*model.ModelMeta, error) { - if result, err := m.DO.Take(); err != nil { - return nil, err - } else { - return result.(*model.ModelMeta), nil - } -} - -func (m modelMetaDo) Last() (*model.ModelMeta, error) { - if result, err := m.DO.Last(); err != nil { - return nil, err - } else { - return result.(*model.ModelMeta), nil - } -} - -func (m modelMetaDo) Find() ([]*model.ModelMeta, error) { - result, err := m.DO.Find() - return result.([]*model.ModelMeta), err -} - -func (m modelMetaDo) FindInBatch(batchSize int, fc func(tx gen.Dao, batch int) error) (results []*model.ModelMeta, err error) { - buf := make([]*model.ModelMeta, 0, batchSize) - err = m.DO.FindInBatches(&buf, batchSize, func(tx gen.Dao, batch int) error { - defer func() { results = append(results, buf...) }() - return fc(tx, batch) - }) - return results, err -} - -func (m modelMetaDo) FindInBatches(result *[]*model.ModelMeta, batchSize int, fc func(tx gen.Dao, batch int) error) error { - return m.DO.FindInBatches(result, batchSize, fc) -} - -func (m modelMetaDo) Attrs(attrs ...field.AssignExpr) IModelMetaDo { - return m.withDO(m.DO.Attrs(attrs...)) -} - -func (m modelMetaDo) Assign(attrs ...field.AssignExpr) IModelMetaDo { - return m.withDO(m.DO.Assign(attrs...)) -} - -func (m modelMetaDo) Joins(fields ...field.RelationField) IModelMetaDo { - for _, _f := range fields { - m = *m.withDO(m.DO.Joins(_f)) - } - return &m -} - -func (m modelMetaDo) Preload(fields ...field.RelationField) IModelMetaDo { - for _, _f := range fields { - m = *m.withDO(m.DO.Preload(_f)) - } - return &m -} - -func (m modelMetaDo) FirstOrInit() (*model.ModelMeta, error) { - if result, err := m.DO.FirstOrInit(); err != nil { - return nil, err - } else { - return result.(*model.ModelMeta), nil - } -} - -func (m modelMetaDo) FirstOrCreate() (*model.ModelMeta, error) { - if result, err := m.DO.FirstOrCreate(); err != nil { - return nil, err - } else { - return result.(*model.ModelMeta), nil - } -} - -func (m modelMetaDo) FindByPage(offset int, limit int) (result []*model.ModelMeta, count int64, err error) { - result, err = m.Offset(offset).Limit(limit).Find() - if err != nil { - return - } - - if size := len(result); 0 < limit && 0 < size && size < limit { - count = int64(size + offset) - return - } - - count, err = m.Offset(-1).Limit(-1).Count() - return -} - -func (m modelMetaDo) ScanByPage(result interface{}, offset int, limit int) (count int64, err error) { - count, err = m.Count() - if err != nil { - return - } - - err = m.Offset(offset).Limit(limit).Scan(result) - return -} - -func (m modelMetaDo) Scan(result interface{}) (err error) { - return m.DO.Scan(result) -} - -func (m modelMetaDo) Delete(models ...*model.ModelMeta) (result gen.ResultInfo, err error) { - return m.DO.Delete(models) -} - -func (m *modelMetaDo) withDO(do gen.Dao) *modelMetaDo { - m.DO = *do.(*gen.DO) - return m -} diff --git a/backend/domain/modelmgr/service/model_manager.go b/backend/domain/modelmgr/service/model_manager.go deleted file mode 100644 index 3839cccf..00000000 --- a/backend/domain/modelmgr/service/model_manager.go +++ /dev/null @@ -1,389 +0,0 @@ -/* - * Copyright 2025 coze-dev Authors - * - * Licensed under the Apache License, Version 2.0 (the "License"); - * you may not use this file except in compliance with the License. - * You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package service - -import ( - "context" - "encoding/json" - "fmt" - "time" - - "gorm.io/gorm" - - modelmgrModel "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr/entity" - "github.com/coze-dev/coze-studio/backend/domain/modelmgr/internal/dal/dao" - dmodel "github.com/coze-dev/coze-studio/backend/domain/modelmgr/internal/dal/model" - uploadEntity "github.com/coze-dev/coze-studio/backend/domain/upload/entity" - modelcontract "github.com/coze-dev/coze-studio/backend/infra/contract/chatmodel" - "github.com/coze-dev/coze-studio/backend/infra/contract/idgen" - "github.com/coze-dev/coze-studio/backend/infra/contract/storage" - "github.com/coze-dev/coze-studio/backend/pkg/lang/slices" -) - -func NewModelManager(db *gorm.DB, idgen idgen.IDGenerator, oss storage.Storage) modelmgr.Manager { - return &modelManager{ - idgen: idgen, - oss: oss, - modelMetaRepo: dao.NewModelMetaDAO(db), - modelEntityRepo: dao.NewModelEntityDAO(db), - } -} - -type modelManager struct { - idgen idgen.IDGenerator - oss storage.Storage - - modelMetaRepo dao.ModelMetaRepo - modelEntityRepo dao.ModelEntityRepo -} - -func (m *modelManager) CreateModelMeta(ctx context.Context, meta *entity.ModelMeta) (resp *entity.ModelMeta, err error) { - if err = m.alignProtocol(meta); err != nil { - return nil, err - } - - id := meta.ID - if id == 0 { - id, err = m.idgen.GenID(ctx) - if err != nil { - return nil, err - } - } - - desc, err := json.Marshal(meta.Description) - if err != nil { - return nil, err - } - - now := time.Now().UnixMilli() - if err = m.modelMetaRepo.Create(ctx, &dmodel.ModelMeta{ - ID: id, - ModelName: meta.Name, - Protocol: string(meta.Protocol), - IconURI: meta.IconURI, - IconURL: meta.IconURL, - Capability: meta.Capability, - ConnConfig: meta.ConnConfig, - Status: meta.Status, - Description: string(desc), - CreatedAt: now, - UpdatedAt: now, - DeletedAt: gorm.DeletedAt{}, - }); err != nil { - return nil, err - } - - return &entity.ModelMeta{ - ID: id, - Name: meta.Name, - Description: meta.Description, - CreatedAtMs: now, - UpdatedAtMs: now, - - Protocol: meta.Protocol, - Capability: meta.Capability, - ConnConfig: meta.ConnConfig, - Status: meta.Status, - }, nil -} - -func (m *modelManager) UpdateModelMetaStatus(ctx context.Context, id int64, status entity.ModelMetaStatus) error { - return m.modelMetaRepo.UpdateStatus(ctx, id, status) -} - -func (m *modelManager) DeleteModelMeta(ctx context.Context, id int64) error { - return m.modelMetaRepo.Delete(ctx, id) -} - -func (m *modelManager) ListModelMeta(ctx context.Context, req *modelmgr.ListModelMetaRequest) (*modelmgr.ListModelMetaResponse, error) { - status := req.Status - if len(status) == 0 { - status = []entity.ModelMetaStatus{modelmgrModel.StatusInUse} - } - - pos, next, hasMore, err := m.modelMetaRepo.List(ctx, req.FuzzyModelName, status, req.Limit, req.Cursor) - if err != nil { - return nil, err - } - - dos, err := m.fromModelMetaPOs(ctx, pos) - if err != nil { - return nil, err - } - - return &modelmgr.ListModelMetaResponse{ - ModelMetaList: dos, - HasMore: hasMore, - NextCursor: next, - }, nil -} - -func (m *modelManager) MGetModelMetaByID(ctx context.Context, req *modelmgr.MGetModelMetaRequest) ([]*entity.ModelMeta, error) { - if len(req.IDs) == 0 { - return nil, nil - } - - pos, err := m.modelMetaRepo.MGetByID(ctx, req.IDs) - if err != nil { - return nil, err - } - - dos, err := m.fromModelMetaPOs(ctx, pos) - if err != nil { - return nil, err - } - - return dos, nil -} - -func (m *modelManager) CreateModel(ctx context.Context, e *entity.Model) (*entity.Model, error) { - // check if meta id exists - metaPO, err := m.modelMetaRepo.GetByID(ctx, e.Meta.ID) - if err != nil { - return nil, err - } - if metaPO == nil { - return nil, fmt.Errorf("[CreateModel] mode meta not found, model_meta id=%d", e.Meta.ID) - } - id := e.ID - if id == 0 { - id, err = m.idgen.GenID(ctx) - if err != nil { - return nil, err - } - } - - now := time.Now().UnixMilli() - // TODO(@fanlv) : do -> po 放到 dal 里面去 - if err = m.modelEntityRepo.Create(ctx, &dmodel.ModelEntity{ - ID: id, - MetaID: e.Meta.ID, - Name: e.Name, - Description: e.Description, - DefaultParams: e.DefaultParameters, - Status: modelmgrModel.ModelEntityStatusInUse, - CreatedAt: now, - UpdatedAt: now, - }); err != nil { - return nil, err - } - - resp := &entity.Model{ - Model: &modelmgrModel.Model{ - ID: id, - Name: e.Name, - CreatedAtMs: now, - UpdatedAtMs: now, - Meta: e.Meta, - }, - } - - return resp, nil -} - -func (m *modelManager) DeleteModel(ctx context.Context, id int64) error { - return m.modelEntityRepo.Delete(ctx, id) -} - -func (m *modelManager) ListModel(ctx context.Context, req *modelmgr.ListModelRequest) (*modelmgr.ListModelResponse, error) { - var sc *int64 - - status := req.Status - if len(status) == 0 { - status = []modelmgrModel.ModelEntityStatus{modelmgrModel.ModelEntityStatusDefault, modelmgrModel.ModelEntityStatusInUse} - } - - pos, next, hasMore, err := m.modelEntityRepo.List(ctx, req.FuzzyModelName, sc, status, req.Limit, req.Cursor) - if err != nil { - return nil, err - } - - pos = moveDefaultModelToFirst(pos) - resp, err := m.fromModelPOs(ctx, pos) - if err != nil { - return nil, err - } - - return &modelmgr.ListModelResponse{ - ModelList: resp, - HasMore: hasMore, - NextCursor: next, - }, nil -} - -func (m *modelManager) MGetModelByID(ctx context.Context, req *modelmgr.MGetModelRequest) ([]*entity.Model, error) { - if len(req.IDs) == 0 { - return nil, nil - } - - pos, err := m.modelEntityRepo.MGet(ctx, req.IDs) - if err != nil { - return nil, err - } - - resp, err := m.fromModelPOs(ctx, pos) - if err != nil { - return nil, err - } - - return resp, nil -} - -func (m *modelManager) alignProtocol(meta *entity.ModelMeta) error { - if meta.Protocol == "" { - return fmt.Errorf("protocol not provided") - } - - config := meta.ConnConfig - if config == nil { - return fmt.Errorf("ConnConfig not provided, protocol=%s", meta.Protocol) - } - - return nil -} - -func (m *modelManager) fromModelMetaPOs(ctx context.Context, pos []*dmodel.ModelMeta) ([]*entity.ModelMeta, error) { - uris := make(map[string]string) - - for _, po := range pos { - if po == nil || po.IconURL != "" { - continue - } - if po.IconURI == "" { - po.IconURI = uploadEntity.ModelIconURI - } - uris[po.IconURI] = "" - } - - for uri := range uris { - url, err := m.oss.GetObjectUrl(ctx, uri) - if err != nil { - return nil, err - } - uris[uri] = url - } - - dos, err := slices.TransformWithErrorCheck(pos, func(po *dmodel.ModelMeta) (*entity.ModelMeta, error) { - if po == nil { - return nil, nil - } - url := po.IconURL - if url == "" { - url = uris[po.IconURI] - } - - desc := &modelmgrModel.MultilingualText{} - if unmarshalErr := json.Unmarshal([]byte(po.Description), desc); unmarshalErr != nil { - return nil, unmarshalErr - } - - return &entity.ModelMeta{ - ID: po.ID, - Name: po.ModelName, - IconURI: po.IconURI, - IconURL: url, - - Description: desc, - CreatedAtMs: po.CreatedAt, - UpdatedAtMs: po.UpdatedAt, - DeletedAtMs: po.DeletedAt.Time.UnixMilli(), - - Protocol: modelcontract.Protocol(po.Protocol), - Capability: po.Capability, - ConnConfig: po.ConnConfig, - Status: po.Status, - }, nil - }) - if err != nil { - return nil, err - } - - return dos, nil -} - -func (m *modelManager) fromModelPOs(ctx context.Context, pos []*dmodel.ModelEntity) ([]*entity.Model, error) { - if len(pos) == 0 { - return nil, nil - } - - resp := make([]*entity.Model, 0, len(pos)) - metaIDSet := make(map[int64]struct{}) - for _, po := range pos { - resp = append(resp, &entity.Model{ - Model: &modelmgrModel.Model{ - ID: po.ID, - Name: po.Name, - Description: po.Description, - DefaultParameters: po.DefaultParams, - CreatedAtMs: po.CreatedAt, - UpdatedAtMs: po.UpdatedAt, - Meta: entity.ModelMeta{ - ID: po.MetaID, - }, - }, - }) - metaIDSet[po.MetaID] = struct{}{} - } - - metaIDSlice := make([]int64, 0, len(metaIDSet)) - for id := range metaIDSet { - metaIDSlice = append(metaIDSlice, id) - } - - modelMetaSlice, err := m.MGetModelMetaByID(ctx, &modelmgr.MGetModelMetaRequest{IDs: metaIDSlice}) - if err != nil { - return nil, err - } - - metaID2Meta := make(map[int64]*entity.ModelMeta) - for i := range modelMetaSlice { - item := modelMetaSlice[i] - if item.IconURL == "" { - url, err := m.oss.GetObjectUrl(ctx, item.IconURI) - if err != nil { - return nil, err - } - item.IconURL = url - } - metaID2Meta[item.ID] = item - } - - for _, r := range resp { - meta, found := metaID2Meta[r.Meta.ID] - if !found { - return nil, fmt.Errorf("[ListModel] model meta not found, model_entity id=%v, model_meta id=%v", r.ID, r.Meta.ID) - } - r.Meta = *meta - } - - return resp, nil -} - -func moveDefaultModelToFirst(ms []*dmodel.ModelEntity) []*dmodel.ModelEntity { - orders := make([]*dmodel.ModelEntity, len(ms)) - copy(orders, ms) - - for i, m := range orders { - if i != 0 && m.Status == modelmgrModel.ModelEntityStatusDefault { - orders[0], orders[i] = orders[i], orders[0] - break - } - } - return orders -} diff --git a/backend/domain/modelmgr/service/model_manager_with_cache.go b/backend/domain/modelmgr/service/model_manager_with_cache.go deleted file mode 100644 index 97a06f09..00000000 --- a/backend/domain/modelmgr/service/model_manager_with_cache.go +++ /dev/null @@ -1,41 +0,0 @@ -/* - * Copyright 2025 coze-dev Authors - * - * Licensed under the Apache License, Version 2.0 (the "License"); - * you may not use this file except in compliance with the License. - * You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package service - -// TODO: 考虑到 model manager 被外部高频读+运行，修改/删除频率很低，基本没有实时更新需求，可进行 cache -// 1. model_meta -// 2. model_entity -// 3. ChatModel - -// func (m *modelManager) buildOptions(req *model.ChatRequest) []cm.Option { -// var opts []cm.Option -// -// if len(req.Tools) > 0 { -// opts = append(opts, cm.WithTools(req.Tools)) -// } -// if req.Temperature != nil { -// opts = append(opts, cm.WithTemperature(float32(*req.Temperature))) -// } -// if req.MaxTokens != nil { -// opts = append(opts, cm.WithMaxTokens(*req.MaxTokens)) -// } -// if req.TopP != nil { -// opts = append(opts, cm.WithTopP(float32(*req.TopP))) -// } -// // TODO: support frequency_penalty, presence_penalty, top_k -// return opts -//} diff --git a/backend/domain/workflow/crossdomain/model/model.go b/backend/domain/workflow/crossdomain/model/model.go index edabef35..aef868c0 100644 --- a/backend/domain/workflow/crossdomain/model/model.go +++ b/backend/domain/workflow/crossdomain/model/model.go @@ -20,8 +20,7 @@ import ( "context" "github.com/cloudwego/eino/components/model" - - "github.com/coze-dev/coze-studio/backend/crossdomain/contract/crossmodelmgr" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" ) type LLMParams struct { @@ -59,5 +58,5 @@ func SetManager(m Manager) { //go:generate mockgen -destination modelmock/model_mock.go --package mockmodel -source model.go type Manager interface { - GetModel(ctx context.Context, params *LLMParams) (model.BaseChatModel, *crossmodelmgr.Model, error) + GetModel(ctx context.Context, params *LLMParams) (model.BaseChatModel, *modelmgr.Model, error) } diff --git a/backend/domain/workflow/crossdomain/model/modelmock/model_mock.go b/backend/domain/workflow/crossdomain/model/modelmock/model_mock.go index 13c061da..438825cf 100644 --- a/backend/domain/workflow/crossdomain/model/modelmock/model_mock.go +++ b/backend/domain/workflow/crossdomain/model/modelmock/model_mock.go @@ -1,19 +1,3 @@ -/* - * Copyright 2025 coze-dev Authors - * - * Licensed under the Apache License, Version 2.0 (the "License"); - * you may not use this file except in compliance with the License. - * You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - // Code generated by MockGen. DO NOT EDIT. // Source: model.go // @@ -29,9 +13,9 @@ import ( context "context" reflect "reflect" - crossmodelmgr "github.com/coze-dev/coze-studio/backend/crossdomain/contract/crossmodelmgr" - model "github.com/coze-dev/coze-studio/backend/domain/workflow/crossdomain/model" - model0 "github.com/cloudwego/eino/components/model" + model "github.com/cloudwego/eino/components/model" + model0 "github.com/coze-dev/coze-studio/backend/domain/workflow/crossdomain/model" + modelmgr "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" gomock "go.uber.org/mock/gomock" ) @@ -39,7 +23,6 @@ import ( type MockManager struct { ctrl *gomock.Controller recorder *MockManagerMockRecorder - isgomock struct{} } // MockManagerMockRecorder is the mock recorder for MockManager. @@ -60,11 +43,11 @@ func (m *MockManager) EXPECT() *MockManagerMockRecorder { } // GetModel mocks base method. -func (m *MockManager) GetModel(ctx context.Context, params *model.LLMParams) (model0.BaseChatModel, *crossmodelmgr.Model, error) { +func (m *MockManager) GetModel(ctx context.Context, params *model0.LLMParams) (model.BaseChatModel, *modelmgr.Model, error) { m.ctrl.T.Helper() ret := m.ctrl.Call(m, "GetModel", ctx, params) - ret0, _ := ret[0].(model0.BaseChatModel) - ret1, _ := ret[1].(*crossmodelmgr.Model) + ret0, _ := ret[0].(model.BaseChatModel) + ret1, _ := ret[1].(*modelmgr.Model) ret2, _ := ret[2].(error) return ret0, ret1, ret2 } diff --git a/backend/domain/workflow/internal/compose/test/llm_test.go b/backend/domain/workflow/internal/compose/test/llm_test.go index 8c37c07d..577c4aa0 100644 --- a/backend/domain/workflow/internal/compose/test/llm_test.go +++ b/backend/domain/workflow/internal/compose/test/llm_test.go @@ -31,10 +31,10 @@ import ( model2 "github.com/cloudwego/eino/components/model" "github.com/cloudwego/eino/compose" "github.com/cloudwego/eino/schema" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" "github.com/stretchr/testify/assert" "go.uber.org/mock/gomock" - "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" "github.com/coze-dev/coze-studio/backend/domain/workflow/crossdomain/model" mockmodel "github.com/coze-dev/coze-studio/backend/domain/workflow/crossdomain/model/modelmock" "github.com/coze-dev/coze-studio/backend/domain/workflow/entity" diff --git a/backend/domain/workflow/internal/compose/to_node.go b/backend/domain/workflow/internal/compose/to_node.go index f7b79274..09f60c83 100644 --- a/backend/domain/workflow/internal/compose/to_node.go +++ b/backend/domain/workflow/internal/compose/to_node.go @@ -28,8 +28,8 @@ import ( "github.com/cloudwego/eino/components/tool" "github.com/cloudwego/eino/compose" "github.com/cloudwego/eino/schema" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" - crossmodelmgr "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" workflow3 "github.com/coze-dev/coze-studio/backend/api/model/ocean/cloud/workflow" workflow2 "github.com/coze-dev/coze-studio/backend/domain/workflow" crosscode "github.com/coze-dev/coze-studio/backend/domain/workflow/crossdomain/code" @@ -91,7 +91,7 @@ func (s *NodeSchema) ToLLMConfig(ctx context.Context) (*llm.Config, error) { var ( err error chatModel, fallbackM einomodel.BaseChatModel - info, fallbackI *crossmodelmgr.Model + info, fallbackI *modelmgr.Model modelWithInfo llm.ModelWithInfo ) diff --git a/backend/domain/workflow/internal/nodes/llm/model_with_info.go b/backend/domain/workflow/internal/nodes/llm/model_with_info.go index eaa6ba2f..fbcb3a59 100644 --- a/backend/domain/workflow/internal/nodes/llm/model_with_info.go +++ b/backend/domain/workflow/internal/nodes/llm/model_with_info.go @@ -24,28 +24,27 @@ import ( "github.com/cloudwego/eino/components" "github.com/cloudwego/eino/components/model" "github.com/cloudwego/eino/schema" - - crossmodelmgr "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" "github.com/coze-dev/coze-studio/backend/domain/workflow/internal/execute" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" ) type ModelWithInfo interface { model.BaseChatModel - Info(ctx context.Context) *crossmodelmgr.Model + Info(ctx context.Context) *modelmgr.Model } type ModelForLLM struct { Model model.BaseChatModel - MInfo *crossmodelmgr.Model + MInfo *modelmgr.Model FallbackModel model.BaseChatModel - FallbackInfo *crossmodelmgr.Model + FallbackInfo *modelmgr.Model UseFallback func(ctx context.Context) bool modelEnableCallback bool fallbackEnableCallback bool } -func NewModel(m model.BaseChatModel, info *crossmodelmgr.Model) *ModelForLLM { +func NewModel(m model.BaseChatModel, info *modelmgr.Model) *ModelForLLM { return &ModelForLLM{ Model: m, MInfo: info, @@ -57,7 +56,7 @@ func NewModel(m model.BaseChatModel, info *crossmodelmgr.Model) *ModelForLLM { } } -func NewModelWithFallback(m, f model.BaseChatModel, info, fInfo *crossmodelmgr.Model) *ModelForLLM { +func NewModelWithFallback(m, f model.BaseChatModel, info, fInfo *modelmgr.Model) *ModelForLLM { return &ModelForLLM{ Model: m, MInfo: info, @@ -175,7 +174,7 @@ func (m *ModelForLLM) IsCallbacksEnabled() bool { return true } -func (m *ModelForLLM) Info(ctx context.Context) *crossmodelmgr.Model { +func (m *ModelForLLM) Info(ctx context.Context) *modelmgr.Model { if m.UseFallback(ctx) { return m.FallbackInfo } diff --git a/backend/domain/workflow/internal/nodes/llm/prompt.go b/backend/domain/workflow/internal/nodes/llm/prompt.go index 4ca39579..7b4459cc 100644 --- a/backend/domain/workflow/internal/nodes/llm/prompt.go +++ b/backend/domain/workflow/internal/nodes/llm/prompt.go @@ -22,8 +22,8 @@ import ( "github.com/cloudwego/eino/components/prompt" "github.com/cloudwego/eino/schema" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" - "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" "github.com/coze-dev/coze-studio/backend/domain/workflow/entity/vo" "github.com/coze-dev/coze-studio/backend/domain/workflow/internal/execute" "github.com/coze-dev/coze-studio/backend/domain/workflow/internal/nodes" diff --git a/backend/infra/contract/chatmodel/config.go b/backend/infra/contract/chatmodel/config.go index 08f4cc3a..a8c49580 100644 --- a/backend/infra/contract/chatmodel/config.go +++ b/backend/infra/contract/chatmodel/config.go @@ -25,9 +25,9 @@ import ( ) type Config struct { - BaseURL string `json:"base_url,omitempty" yaml:"base_url,omitempty"` - APIKey string `json:"api_key,omitempty" yaml:"api_key,omitempty"` - Timeout time.Duration `json:"timeout,omitempty" yaml:"timeout,omitempty"` + BaseURL string `json:"base_url,omitempty" yaml:"base_url"` + APIKey string `json:"api_key,omitempty" yaml:"api_key"` + Timeout time.Duration `json:"timeout,omitempty" yaml:"timeout"` Model string `json:"model" yaml:"model"` Temperature *float32 `json:"temperature,omitempty" yaml:"temperature,omitempty"` diff --git a/backend/api/model/crossdomain/modelmgr/const.go b/backend/infra/contract/modelmgr/const.go similarity index 69% rename from backend/api/model/crossdomain/modelmgr/const.go rename to backend/infra/contract/modelmgr/const.go index 2e1150e6..c9f1383b 100644 --- a/backend/api/model/crossdomain/modelmgr/const.go +++ b/backend/infra/contract/modelmgr/const.go @@ -43,12 +43,13 @@ const ( ModalVideo Modal = "video" ) -type ModelMetaStatus int64 // 模型实体状态 +type ModelStatus int64 const ( - StatusInUse ModelMetaStatus = 1 // 应用中，可使用可新建 - StatusPending ModelMetaStatus = 5 // 待下线，可使用不可新建 - StatusDeleted ModelMetaStatus = 10 // 已下线，不可使用不可新建 + StatusDefault ModelStatus = 0 // 未配置时的默认状态，表现等同 StatusInUse + StatusInUse ModelStatus = 1 // 应用中，可使用可新建 + StatusPending ModelStatus = 5 // 待下线，可使用不可新建 + StatusDeleted ModelStatus = 10 // 已下线，不可使用不可新建 ) type Widget string @@ -57,12 +58,3 @@ const ( WidgetSlider Widget = "slider" WidgetRadioButtons Widget = "radio_buttons" ) - -type ModelEntityStatus int64 - -const ( - ModelEntityStatusDefault ModelEntityStatus = 0 - ModelEntityStatusInUse ModelEntityStatus = 1 - ModelEntityStatusPending ModelEntityStatus = 5 - ModelEntityStatusDeleted ModelEntityStatus = 10 -) diff --git a/backend/api/model/crossdomain/modelmgr/modelmgr.go b/backend/infra/contract/modelmgr/desc.go similarity index 86% rename from backend/api/model/crossdomain/modelmgr/modelmgr.go rename to backend/infra/contract/modelmgr/desc.go index f0dc79c3..2bceff0d 100644 --- a/backend/api/model/crossdomain/modelmgr/modelmgr.go +++ b/backend/infra/contract/modelmgr/desc.go @@ -8,21 +8,28 @@ import ( "github.com/coze-dev/coze-studio/backend/pkg/i18n" ) -type MGetModelRequest struct { - IDs []int64 +type Model struct { + ID int64 `yaml:"id"` + Name string `yaml:"name"` + IconURI string `yaml:"icon_uri"` + IconURL string `yaml:"icon_url"` + Description *MultilingualText `yaml:"description"` + DefaultParameters []*Parameter `yaml:"default_parameters"` + Meta ModelMeta `yaml:"meta"` } -type Model struct { - ID int64 `yaml:"id"` - Name string `yaml:"name"` - Description string `yaml:"description"` - DefaultParameters []*Parameter `yaml:"default_parameters"` +func (m *Model) FindParameter(name ParameterName) (*Parameter, bool) { + if len(m.DefaultParameters) == 0 { + return nil, false + } - CreatedAtMs int64 - UpdatedAtMs int64 - DeletedAtMs int64 + for _, param := range m.DefaultParameters { + if param.Name == name { + return param, true + } + } - Meta ModelMeta `yaml:"meta"` + return nil, false } type Parameter struct { @@ -100,20 +107,11 @@ func (p *Parameter) GetString(tp DefaultType) (string, error) { } type ModelMeta struct { - ID int64 `yaml:"id"` - Name string `yaml:"name"` - IconURI string `yaml:"icon_uri"` - IconURL string `yaml:"icon_url"` - Description *MultilingualText `yaml:"description"` - - CreatedAtMs int64 - UpdatedAtMs int64 - DeletedAtMs int64 - + Name string `yaml:"name"` Protocol chatmodel.Protocol `yaml:"protocol"` // 模型通信协议 Capability *Capability `yaml:"capability"` // 模型能力 ConnConfig *chatmodel.Config `yaml:"conn_config"` // 模型连接配置 - Status ModelMetaStatus `yaml:"status"` // 模型状态 + Status ModelStatus `yaml:"status"` // 模型状态 } type DefaultValue map[DefaultType]string diff --git a/backend/infra/contract/modelmgr/modelmgr.go b/backend/infra/contract/modelmgr/modelmgr.go new file mode 100644 index 00000000..9e32cde3 --- /dev/null +++ b/backend/infra/contract/modelmgr/modelmgr.go @@ -0,0 +1,27 @@ +package modelmgr + +import ( + "context" +) + +type Manager interface { + ListModel(ctx context.Context, req *ListModelRequest) (*ListModelResponse, error) + MGetModelByID(ctx context.Context, req *MGetModelRequest) ([]*Model, error) +} + +type ListModelRequest struct { + FuzzyModelName *string + Status []ModelStatus // default is default and in_use status + Limit int + Cursor *string +} + +type ListModelResponse struct { + ModelList []*Model + HasMore bool + NextCursor *string +} + +type MGetModelRequest struct { + IDs []int64 +} diff --git a/backend/infra/impl/modelmgr/static/modelmgr.go b/backend/infra/impl/modelmgr/static/modelmgr.go new file mode 100644 index 00000000..e27ba756 --- /dev/null +++ b/backend/infra/impl/modelmgr/static/modelmgr.go @@ -0,0 +1,80 @@ +package static + +import ( + "context" + "strconv" + "strings" + + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" + "github.com/coze-dev/coze-studio/backend/pkg/lang/ptr" + "github.com/coze-dev/coze-studio/backend/pkg/lang/sets" +) + +func NewModelMgr(staticModels []*modelmgr.Model) (modelmgr.Manager, error) { + mapping := make(map[int64]*modelmgr.Model, len(staticModels)) + for i := range staticModels { + mapping[staticModels[i].ID] = staticModels[i] + } + return &staticModelManager{ + models: staticModels, + modelMapping: mapping, + }, nil +} + +type staticModelManager struct { + models []*modelmgr.Model + modelMapping map[int64]*modelmgr.Model +} + +func (s *staticModelManager) ListModel(_ context.Context, req *modelmgr.ListModelRequest) (*modelmgr.ListModelResponse, error) { + startIdx := 0 + if req.Cursor != nil { + start, err := strconv.ParseInt(*req.Cursor, 10, 64) + if err != nil { + return nil, err + } + startIdx = int(start) + } + + limit := req.Limit + if limit == 0 { + limit = 100 + } + + var ( + i int + respList []*modelmgr.Model + statSet = sets.FromSlice(req.Status) + ) + + for i = startIdx; i < len(s.models) && len(respList) < limit; i++ { + m := s.models[i] + if req.FuzzyModelName != nil && !strings.Contains(m.Name, *req.FuzzyModelName) { + continue + } + if len(statSet) > 0 && !statSet.Contains(m.Meta.Status) { + continue + } + respList = append(respList, m) + } + + resp := &modelmgr.ListModelResponse{ + ModelList: respList, + } + resp.HasMore = i != len(s.models) + if resp.HasMore { + resp.NextCursor = ptr.Of(strconv.FormatInt(int64(i), 10)) + } + + return resp, nil +} + +func (s *staticModelManager) MGetModelByID(_ context.Context, req *modelmgr.MGetModelRequest) ([]*modelmgr.Model, error) { + resp := make([]*modelmgr.Model, 0, len(s.models)) + for _, id := range req.IDs { + if m, found := s.modelMapping[id]; found { + resp = append(resp, m) + } + } + return resp, nil +} diff --git a/backend/internal/mock/domain/agent/singleagent/model_mgr_mock.go b/backend/internal/mock/domain/agent/singleagent/model_mgr_mock.go index 2d9f81e3..4bbd2266 100644 --- a/backend/internal/mock/domain/agent/singleagent/model_mgr_mock.go +++ b/backend/internal/mock/domain/agent/singleagent/model_mgr_mock.go @@ -1,9 +1,9 @@ // Code generated by MockGen. DO NOT EDIT. -// Source: model_manager.go +// Source: backend/infra/contract/modelmgr/modelmgr.go // // Generated by this command: // -// mockgen -destination ../../../../internal/mock/domain/agent/singleagent/model_mgr_mock.go --package mock -source model_manager.go +// mockgen --destination backend/internal/mock/domain/agent/singleagent/model_mgr_mock.go --package mock -source backend/infra/contract/modelmgr/modelmgr.go // // Package mock is a generated GoMock package. @@ -13,47 +13,59 @@ import ( context "context" reflect "reflect" + modelmgr "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" gomock "go.uber.org/mock/gomock" - - modelmgr "github.com/coze-dev/coze-studio/backend/domain/modelmgr" - entity "github.com/coze-dev/coze-studio/backend/domain/modelmgr/entity" ) -// MockModelMgr is a mock of ModelMgr interface. -type MockModelMgr struct { +// MockManager is a mock of Manager interface. +type MockManager struct { ctrl *gomock.Controller - recorder *MockModelMgrMockRecorder - isgomock struct{} + recorder *MockManagerMockRecorder } -// MockModelMgrMockRecorder is the mock recorder for MockModelMgr. -type MockModelMgrMockRecorder struct { - mock *MockModelMgr +// MockManagerMockRecorder is the mock recorder for MockManager. +type MockManagerMockRecorder struct { + mock *MockManager } -// NewMockModelMgr creates a new mock instance. -func NewMockModelMgr(ctrl *gomock.Controller) *MockModelMgr { - mock := &MockModelMgr{ctrl: ctrl} - mock.recorder = &MockModelMgrMockRecorder{mock} +// NewMockManager creates a new mock instance. +func NewMockManager(ctrl *gomock.Controller) *MockManager { + mock := &MockManager{ctrl: ctrl} + mock.recorder = &MockManagerMockRecorder{mock} return mock } // EXPECT returns an object that allows the caller to indicate expected use. -func (m *MockModelMgr) EXPECT() *MockModelMgrMockRecorder { +func (m *MockManager) EXPECT() *MockManagerMockRecorder { return m.recorder } +// ListModel mocks base method. +func (m *MockManager) ListModel(ctx context.Context, req *modelmgr.ListModelRequest) (*modelmgr.ListModelResponse, error) { + m.ctrl.T.Helper() + ret := m.ctrl.Call(m, "ListModel", ctx, req) + ret0, _ := ret[0].(*modelmgr.ListModelResponse) + ret1, _ := ret[1].(error) + return ret0, ret1 +} + +// ListModel indicates an expected call of ListModel. +func (mr *MockManagerMockRecorder) ListModel(ctx, req any) *gomock.Call { + mr.mock.ctrl.T.Helper() + return mr.mock.ctrl.RecordCallWithMethodType(mr.mock, "ListModel", reflect.TypeOf((*MockManager)(nil).ListModel), ctx, req) +} + // MGetModelByID mocks base method. -func (m *MockModelMgr) MGetModelByID(ctx context.Context, req *modelmgr.MGetModelRequest) ([]*entity.Model, error) { +func (m *MockManager) MGetModelByID(ctx context.Context, req *modelmgr.MGetModelRequest) ([]*modelmgr.Model, error) { m.ctrl.T.Helper() ret := m.ctrl.Call(m, "MGetModelByID", ctx, req) - ret0, _ := ret[0].([]*entity.Model) + ret0, _ := ret[0].([]*modelmgr.Model) ret1, _ := ret[1].(error) return ret0, ret1 } // MGetModelByID indicates an expected call of MGetModelByID. -func (mr *MockModelMgrMockRecorder) MGetModelByID(ctx, req any) *gomock.Call { +func (mr *MockManagerMockRecorder) MGetModelByID(ctx, req any) *gomock.Call { mr.mock.ctrl.T.Helper() - return mr.mock.ctrl.RecordCallWithMethodType(mr.mock, "MGetModelByID", reflect.TypeOf((*MockModelMgr)(nil).MGetModelByID), ctx, req) + return mr.mock.ctrl.RecordCallWithMethodType(mr.mock, "MGetModelByID", reflect.TypeOf((*MockManager)(nil).MGetModelByID), ctx, req) } diff --git a/backend/internal/testutil/chat_model.go b/backend/internal/testutil/chat_model.go index 8b5650b9..49899f83 100644 --- a/backend/internal/testutil/chat_model.go +++ b/backend/internal/testutil/chat_model.go @@ -27,7 +27,7 @@ import ( "github.com/cloudwego/eino/components/model" "github.com/cloudwego/eino/schema" - "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" + "github.com/coze-dev/coze-studio/backend/infra/contract/modelmgr" "github.com/coze-dev/coze-studio/backend/pkg/logs" ) diff --git a/backend/types/ddl/gen_orm_query.go b/backend/types/ddl/gen_orm_query.go index 5fc2b70e..549e0230 100644 --- a/backend/types/ddl/gen_orm_query.go +++ b/backend/types/ddl/gen_orm_query.go @@ -32,13 +32,11 @@ import ( "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/agentrun" "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/database" - modelEntity "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/modelmgr" "github.com/coze-dev/coze-studio/backend/api/model/crossdomain/plugin" "github.com/coze-dev/coze-studio/backend/api/model/ocean/cloud/bot_common" "github.com/coze-dev/coze-studio/backend/api/model/ocean/cloud/playground" appEntity "github.com/coze-dev/coze-studio/backend/domain/app/entity" variableEntity "github.com/coze-dev/coze-studio/backend/domain/memory/variables/entity" - "github.com/coze-dev/coze-studio/backend/infra/contract/chatmodel" ) var path2Table2Columns2Model = map[string]map[string]map[string]any{ @@ -135,18 +133,18 @@ var path2Table2Columns2Model = map[string]map[string]map[string]any{ }, "variable_instance": {}, }, - "domain/modelmgr/internal/dal/query": { - "model_meta": { - "capability": &modelEntity.Capability{}, - "conn_config": &chatmodel.Config{}, - "status": modelEntity.ModelMetaStatus(0), - }, - "model_entity": { - "scenario": modelEntity.Scenario(0), - "default_params": []*modelEntity.Parameter{}, - "status": modelEntity.ModelEntityStatus(0), - }, - }, + // "domain/modelmgr/internal/dal/query": { + // "model_meta": { + // "capability": &modelmgr.Capability{}, + // "conn_config": &chatmodel.Config{}, + // "status": modelmgr.ModelStatus(0), + // }, + // "model_entity": { + // "scenario": modelmgr.Scenario(0), + // "default_params": []*modelmgr.Parameter{}, + // "status": modelmgr.ModelStatus(0), + // }, + // }, "domain/workflow/internal/repo/dal/query": { "workflow_meta": {}, "workflow_draft": {}, diff --git a/backend/conf/model/icon/claude_v2.png b/docker/volumes/minio/default_icon/claude_v2.png similarity index 100% rename from backend/conf/model/icon/claude_v2.png rename to docker/volumes/minio/default_icon/claude_v2.png diff --git a/backend/conf/model/icon/deepseek_v2.png b/docker/volumes/minio/default_icon/deepseek_v2.png similarity index 100% rename from backend/conf/model/icon/deepseek_v2.png rename to docker/volumes/minio/default_icon/deepseek_v2.png diff --git a/backend/conf/model/icon/doubao_v2.png b/docker/volumes/minio/default_icon/doubao_v2.png similarity index 100% rename from backend/conf/model/icon/doubao_v2.png rename to docker/volumes/minio/default_icon/doubao_v2.png diff --git a/backend/conf/model/icon/gemini_v2.png b/docker/volumes/minio/default_icon/gemini_v2.png similarity index 100% rename from backend/conf/model/icon/gemini_v2.png rename to docker/volumes/minio/default_icon/gemini_v2.png diff --git a/backend/conf/model/icon/ollama.png b/docker/volumes/minio/default_icon/ollama.png similarity index 100% rename from backend/conf/model/icon/ollama.png rename to docker/volumes/minio/default_icon/ollama.png diff --git a/backend/conf/model/icon/openai_v2.png b/docker/volumes/minio/default_icon/openai_v2.png similarity index 100% rename from backend/conf/model/icon/openai_v2.png rename to docker/volumes/minio/default_icon/openai_v2.png diff --git a/backend/conf/model/icon/qwen_v2.png b/docker/volumes/minio/default_icon/qwen_v2.png similarity index 100% rename from backend/conf/model/icon/qwen_v2.png rename to docker/volumes/minio/default_icon/qwen_v2.png