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
This commit is contained in:
徐兆楠
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
@@ -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
|
||||
133
backend/conf/model/template/model_template_ark.yaml
Executable file
133
backend/conf/model/template/model_template_ark.yaml
Executable file
@@ -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
|
||||
108
backend/conf/model/template/model_template_ark_doubao-1.5-lite.yaml
Executable file
108
backend/conf/model/template/model_template_ark_doubao-1.5-lite.yaml
Executable file
@@ -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
|
||||
108
backend/conf/model/template/model_template_ark_doubao-1.5-pro-256k.yaml
Executable file
108
backend/conf/model/template/model_template_ark_doubao-1.5-pro-256k.yaml
Executable file
@@ -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
|
||||
108
backend/conf/model/template/model_template_ark_doubao-1.5-pro-32k.yaml
Executable file
108
backend/conf/model/template/model_template_ark_doubao-1.5-pro-32k.yaml
Executable file
@@ -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
|
||||
131
backend/conf/model/template/model_template_ark_doubao-1.5-thinking-pro.yaml
Executable file
131
backend/conf/model/template/model_template_ark_doubao-1.5-thinking-pro.yaml
Executable file
@@ -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
|
||||
@@ -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
|
||||
131
backend/conf/model/template/model_template_ark_doubao-1.5-vision-lite.yaml
Executable file
131
backend/conf/model/template/model_template_ark_doubao-1.5-vision-lite.yaml
Executable file
@@ -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
|
||||
132
backend/conf/model/template/model_template_ark_doubao-1.5-vision-pro.yaml
Executable file
132
backend/conf/model/template/model_template_ark_doubao-1.5-vision-pro.yaml
Executable file
@@ -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
|
||||
132
backend/conf/model/template/model_template_ark_doubao-seed-1.6-flash.yaml
Executable file
132
backend/conf/model/template/model_template_ark_doubao-seed-1.6-flash.yaml
Executable file
@@ -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
|
||||
132
backend/conf/model/template/model_template_ark_doubao-seed-1.6-thinking.yaml
Executable file
132
backend/conf/model/template/model_template_ark_doubao-seed-1.6-thinking.yaml
Executable file
@@ -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
|
||||
132
backend/conf/model/template/model_template_ark_doubao-seed-1.6.yaml
Executable file
132
backend/conf/model/template/model_template_ark_doubao-seed-1.6.yaml
Executable file
@@ -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
|
||||
108
backend/conf/model/template/model_template_ark_volc_deepseek-r1.yaml
Executable file
108
backend/conf/model/template/model_template_ark_volc_deepseek-r1.yaml
Executable file
@@ -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
|
||||
108
backend/conf/model/template/model_template_ark_volc_deepseek-v3.yaml
Executable file
108
backend/conf/model/template/model_template_ark_volc_deepseek-v3.yaml
Executable file
@@ -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
|
||||
201
backend/conf/model/template/model_template_basic.yaml
Executable file
201
backend/conf/model/template/model_template_basic.yaml
Executable file
@@ -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
|
||||
90
backend/conf/model/template/model_template_claude.yaml
Executable file
90
backend/conf/model/template/model_template_claude.yaml
Executable file
@@ -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
|
||||
107
backend/conf/model/template/model_template_deepseek.yaml
Executable file
107
backend/conf/model/template/model_template_deepseek.yaml
Executable file
@@ -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
|
||||
139
backend/conf/model/template/model_template_gemini.yaml
Executable file
139
backend/conf/model/template/model_template_gemini.yaml
Executable file
@@ -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
|
||||
84
backend/conf/model/template/model_template_ollama.yaml
Executable file
84
backend/conf/model/template/model_template_ollama.yaml
Executable file
@@ -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
|
||||
171
backend/conf/model/template/model_template_openai.yaml
Executable file
171
backend/conf/model/template/model_template_openai.yaml
Executable file
@@ -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
|
||||
106
backend/conf/model/template/model_template_qwen.yaml
Executable file
106
backend/conf/model/template/model_template_qwen.yaml
Executable file
@@ -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
|
||||
Reference in New Issue
Block a user