Channel Management Module¶

Function Description

The API prefix is uniformly http(s)://<your-domain>

HTTPS should be used in production environments to secure authentication tokens. HTTP is only recommended for development environments.

A complete management system for AI service provider channels. Supports CRUD operations for channels, batch operations, connectivity testing, balance inquiry, and tag management. Includes advanced features like model capability synchronization and channel duplication.

🔐 Administrator Authentication¶

Get Channel List¶

Interface Name: Get Channel List
HTTP Method: GET
Path: /api/channel/
Authentication Requirement: Administrator
Function Description: Paginate and retrieve list information for all channels in the system, supporting filtering by type, status, and tag mode

💡 Request Example:

const response = await fetch('/api/channel/?p=1&page_size=20&id_sort=false&tag_mode=false&type=1&status=enabled', {  
  method: 'GET',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  }  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "",  
  "data": {  
    "items": [  
      {  
        "id": 1,  
        "name": "OpenAI渠道",  
        "type": 1,  
        "status": 1,  
        "priority": 10,  
        "weight": 100,  
        "models": "gpt-3.5-turbo,gpt-4",  
        "group": "default",  
        "response_time": 1500,  
        "test_time": 1640995200  
      }  
    ],  
    "total": 50,  
    "type_counts": {  
      "1": 20,  
      "2": 15,  
      "all": 35  
    }  
  }  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "获取渠道列表失败"  
}

🧾 Field Description:

p (Number): Page number, defaults to 1
page_size (Number): Items per page, defaults to 20
id_sort (Boolean): Whether to sort by ID, defaults to sorting by priority
tag_mode (Boolean): Whether to enable tag mode
type (Number): Channel type filter
status (String): Status filter, optional values: "enabled", "disabled", "all"

Search Channels¶

Interface Name: Search Channels
HTTP Method: GET
Path: /api/channel/search
Authentication Requirement: Administrator
Function Description: Search channels based on keywords, group, model, and other conditions

💡 Request Example:

const response = await fetch('/api/channel/search?keyword=openai&group=default&model=gpt-4&id_sort=false&tag_mode=false&p=1&page_size=20&type=1&status=enabled', {  
  method: 'GET',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  }  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "",  
  "data": {  
    "items": [  
      {  
        "id": 1,  
        "name": "OpenAI官方渠道",  
        "type": 1,  
        "status": 1,  
        "models": "gpt-3.5-turbo,gpt-4",  
        "group": "default"  
      }  
    ],  
    "total": 1,  
    "type_counts": {  
      "1": 1,  
      "all": 1  
    }  
  }  
}

❗ Failed Response Example:

{    "success": false,    "message": "搜索渠道失败"  }

🧾 Field Description:

keyword (String): Search keyword, can match channel name
group (String): Group filter condition
model (String): Model filter condition
Other parameters are the same as the Get Channel List interface

Query Channel Model Capabilities¶

Interface Name: Query Channel Model Capabilities
HTTP Method: GET
Path: /api/channel/models
Authentication Requirement: Administrator
Function Description: Retrieve the list of models supported by all channels in the system

💡 Request Example:

const response = await fetch('/api/channel/models', {  
  method: 'GET',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  }  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "",  
  "data": [  
    {  
      "id": "gpt-3.5-turbo",  
      "name": "GPT-3.5 Turbo"  
    },  
    {  
      "id": "gpt-4",  
      "name": "GPT-4"  
    },  
    {  
      "id": "claude-3-sonnet",  
      "name": "Claude 3 Sonnet"  
    }  
  ]  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "获取模型列表失败"  
}

🧾 Field Description:

data (Array): List of model information

id (String): Model ID
name (String): Model display name

Query Enabled Model Capabilities¶

Interface Name: Query Enabled Model Capabilities
HTTP Method: GET
Path: /api/channel/models_enabled
Authentication Requirement: Administrator
Function Description: Retrieve the list of models supported by currently enabled channels

💡 Request Example:

const response = await fetch('/api/channel/models_enabled', {  
  method: 'GET',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  }  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "",  
  "data": [  
    "gpt-3.5-turbo",  
    "gpt-4",  
    "claude-3-sonnet"  
  ]  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "获取启用模型失败"  
}

🧾 Field Description:

data (Array): List of enabled model IDs

Get Single Channel¶

Interface Name: Get Single Channel
HTTP Method: GET
Path: /api/channel/:id
Authentication Requirement: Administrator
Function Description: Retrieve detailed information for the specified channel, excluding sensitive key information

💡 Request Example:

const response = await fetch('/api/channel/123', {  
  method: 'GET',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  }  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "",  
  "data": {  
    "id": 123,  
    "name": "OpenAI渠道",  
    "type": 1,  
    "status": 1,  
    "priority": 10,  
    "weight": 100,  
    "models": "gpt-3.5-turbo,gpt-4",  
    "group": "default",  
    "base_url": "https://api.openai.com",  
    "model_mapping": "{}",  
    "channel_info": {  
      "is_multi_key": false,  
      "multi_key_mode": "random"  
    }  
  }  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "渠道不存在"  
}

🧾 Field Description:

id (Number): Channel ID, passed via URL path
Returns complete channel information, but excludes key fields

Batch Test Channel Connectivity¶

Interface Name: Batch Test Channel Connectivity
HTTP Method: GET
Path: /api/channel/test
Authentication Requirement: Administrator
Function Description: Batch test the connectivity and response time of all or specified channels

💡 Request Example:

const response = await fetch('/api/channel/test?model=gpt-3.5-turbo', {  
  method: 'GET',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  }  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "批量测试完成",  
  "data": {  
    "total": 10,  
    "success": 8,  
    "failed": 2,  
    "results": [  
      {  
        "channel_id": 1,  
        "channel_name": "OpenAI渠道",  
        "success": true,  
        "time": 1.25,  
        "message": ""  
      },  
      {  
        "channel_id": 2,  
        "channel_name": "Claude渠道",  
        "success": false,  
        "time": 0,  
        "message": "连接超时"  
      }  
    ]  
  }  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "批量测试失败"  
}

🧾 Field Description:

model (String): Optional, specifies the test model
results (Array): List of test results
- success (Boolean): Whether the test was successful
- time (Number): Response time (seconds)

Single Channel Test¶

Interface Name: Single Channel Test
HTTP Method: GET
Path: /api/channel/test/:id
Authentication Requirement: Administrator
Function Description: Test the connectivity of the specified channel, supporting a specified test model

💡 Request Example:

const response = await fetch('/api/channel/test/123?model=gpt-4', {  
  method: 'GET',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  }  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "",  
  "time": 1.25  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "API密钥无效",  
  "time": 0.5  
}

🧾 Field Description:

id (Number): Channel ID, passed via URL path
model (String): Optional, specifies the name of the test model
time (Number): Response time (seconds)

Batch Refresh Balance¶

Interface Name: Batch Refresh Balance
HTTP Method: GET
Path: /api/channel/update_balance
Authentication Requirement: Administrator
Function Description: Batch update the balance information for all enabled channels

💡 Request Example:

const response = await fetch('/api/channel/update_balance', {  
  method: 'GET',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  }  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "批量更新余额完成"  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "批量更新余额失败"  
}

🧾 Field Description:

No request parameters, the system automatically updates the balance of all enabled channels

Single Refresh Balance¶

Interface Name: Update Specified Channel Balance
HTTP Method: GET
Path: /api/channel/update_balance/:id
Authentication Requirement: Administrator
Function Description: Update the balance information for the specified channel. Balance inquiry is not supported for multi-key channels.

💡 Request Example:

const response = await fetch('/api/channel/update_balance/123', {  
  method: 'GET',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  }  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "",  
  "balance": 25.50  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "多密钥渠道不支持余额查询"  
}

🧾 Field Description:

id (Number): Channel ID, passed via URL path
balance (Number): Updated channel balance

Add Channel¶

Interface Name: Add Channel
HTTP Method: POST
Path: /api/channel/
Authentication Requirement: Administrator
Function Description: Create a new AI service channel, supporting single, batch, and multi-key modes

💡 Request Example:

const response = await fetch('/api/channel/', {  
  method: 'POST',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  },  
  body: JSON.stringify({  
    mode: "single",  
    channel: {  
      name: "OpenAI渠道",  
      type: 1,  
      key: "<YOUR_API_KEY>",  
      base_url: "https://api.openai.com",  
      models: "gpt-3.5-turbo,gpt-4,claude-3-sonnet",  
      groups: ["default"],  
      priority: 10,  
      weight: 100  
    }  
  })  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": ""  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "不支持的添加模式"  
}

🧾 Field Description:

mode (String): Addition mode, optional values: "single", "batch", "multi_to_single"
multi_key_mode (String): Multi-key mode, required when mode is "multi_to_single"
channel (Object): Channel configuration information
- name (String): Channel name
- type (Number): Channel type
- key (String): API Key
- base_url (String): Base URL
- models (String): Supported model list, comma-separated, optional
- groups (Array): List of available groups
- priority (Number): Priority
- weight (Number): Weight

Update Channel¶

Interface Name: Update Channel
HTTP Method: PUT
Path: /api/channel/
Authentication Requirement: Administrator
Function Description: Update the configuration information of an existing channel

💡 Request Example:

const response = await fetch('/api/channel/', {  
  method: 'PUT',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  },  
  body: JSON.stringify({  
    id: 123,  
    name: "更新的OpenAI渠道",  
    status: 1,  
    priority: 15,  
    weight: 120  
  })  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": ""  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "渠道不存在"  
}

🧾 Field Description:

id (Number): Channel ID, required
Other fields are the same as the Add Channel interface, and are optional

Delete Disabled Channels¶

Interface Name: Delete Disabled Channels
HTTP Method: DELETE
Path: /api/channel/disabled
Authentication Requirement: Administrator
Function Description: Batch delete all disabled channels

💡 Request Example:

const response = await fetch('/api/channel/disabled', {  
  method: 'DELETE',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  }  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "",  
  "data": 5  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "删除失败"  
}

🧾 Field Description:

No request parameters
data (Number): Number of channels deleted

Batch Disable Tagged Channels¶

Interface Name: Batch Disable Tagged Channels
HTTP Method: POST
Path: /api/channel/tag/disabled
Authentication Requirement: Administrator
Function Description: Batch disable channels based on tags

💡 Request Example:

const response = await fetch('/api/channel/tag/disabled', {  
  method: 'POST',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  },  
  body: JSON.stringify({  
    tag: "test-tag"  
  })  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": ""  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "参数错误"  
}

🧾 Field Description:

tag (String): Channel tag to be disabled, required

Batch Enable Tagged Channels¶

Interface Name: Batch Enable Tagged Channels
HTTP Method: POST
Path: /api/channel/tag/enabled
Authentication Requirement: Administrator
Function Description: Batch enable channels based on tags

💡 Request Example:

const response = await fetch('/api/channel/tag/enabled', {  
  method: 'POST',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  },  
  body: JSON.stringify({  
    tag: "production-tag"  
  })  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": ""  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "参数错误"  
}

🧾 Field Description:

tag (String): Channel tag to be enabled, required

Edit Channel Tags¶

Interface Name: Edit Channel Tags
HTTP Method: PUT
Path: /api/channel/tag
Authentication Requirement: Administrator
Function Description: Batch edit channel attributes for specified tags

💡 Request Example:

const response = await fetch('/api/channel/tag', {  
  method: 'PUT',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  },  
  body: JSON.stringify({  
    tag: "old-tag",  
    new_tag: "new-tag",  
    priority: 20,  
    weight: 150,  
    models: "gpt-3.5-turbo,gpt-4,claude-3-sonnet",  
    groups: "default,vip"  
  })  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": ""  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "tag不能为空"  
}

🧾 Field Description:

tag (String): Name of the tag to be edited, required
new_tag (String): New tag name, optional
priority (Number): New priority, optional
weight (Number): New weight, optional
model_mapping (String): Model mapping configuration, optional
models (String): Supported model list, comma-separated, optional
groups (String): List of available groups, comma-separated, optional

Delete Channel¶

Interface Name: Delete Channel
HTTP Method: DELETE
Path: /api/channel/:id
Authentication Requirement: Administrator
Function Description: Hard delete the specified channel. Channel cache will be refreshed after deletion.

💡 Request Example:

const response = await fetch('/api/channel/123', {  
  method: 'DELETE',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  }  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": ""  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "渠道不存在"  
}

🧾 Field Description:

id (Number): Channel ID, passed via URL path

Batch Delete Channels¶

Interface Name: Batch Delete Channels
HTTP Method: POST
Path: /api/channel/batch
Authentication Requirement: Administrator
Function Description: Batch delete channels based on a list of IDs

💡 Request Example:

const response = await fetch('/api/channel/batch', {  
  method: 'POST',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  },  
  body: JSON.stringify({  
    ids: [1, 2, 3, 4, 5]  
  })  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "",  
  "data": 5  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "参数错误"  
}

🧾 Field Description:

ids (Array): List of channel IDs to be deleted, required and cannot be empty
data (Number): Number of channels successfully deleted

Fix Channel Capability Table¶

Interface Name: Fix Channel Capability Table
HTTP Method: POST
Path: /api/channel/fix
Authentication Requirement: Administrator
Function Description: Fix channel capability table data and rebuild the mapping relationship between channels and models

💡 Request Example:

const response = await fetch('/api/channel/fix', {  
  method: 'POST',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  }  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "",  
  "data": {  
    "success": 45,  
    "fails": 2  
  }  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "修复能力表失败"  
}

🧾 Field Description:

No request parameters
data.success (Number): Number of channels successfully fixed
data.fails (Number): Number of channels that failed to fix

Fetch Single Channel Models¶

Interface Name: Fetch Single Channel Models
HTTP Method: GET
Path: /api/channel/fetch_models/:id
Authentication Requirement: Administrator
Function Description: Retrieve the list of available models from the upstream API of the specified channel

💡 Request Example:

const response = await fetch('/api/channel/fetch_models/123', {  
  method: 'GET',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  }  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "",  
  "data": [  
    "gpt-3.5-turbo",  
    "gpt-4",  
    "gpt-4-turbo-preview"  
  ]  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "解析响应失败: invalid character 'H' looking for beginning of value"  
}

🧾 Field Description:

id (Number): Channel ID, passed via URL path
data (Array): List of model IDs fetched from the upstream

Fetch All Channel Models¶

Interface Name: Fetch All Channel Models
HTTP Method: POST
Path: /api/channel/fetch_models
Authentication Requirement: Administrator
Function Description: Retrieve the list of models from the upstream API using the provided configuration information, used for previewing when creating a new channel

💡 Request Example:

const response = await fetch('/api/channel/fetch_models', {  
  method: 'POST',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  },  
  body: JSON.stringify({  
    base_url: "https://api.openai.com",  
    type: 1,  
    key: "<YOUR_API_KEY>"  
  })  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "data": [  
    "gpt-3.5-turbo",  
    "gpt-4",  
    "text-davinci-003"  
  ]  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "Failed to fetch models"  
}

🧾 Field Description:

base_url (String): Base URL, optional, uses default URL if empty
type (Number): Channel type, required
key (String): API Key, required
data (Array): List of models retrieved

Batch Set Channel Tags¶

Interface Name: Batch Set Channel Tags
HTTP Method: POST
Path: /api/channel/batch/tag
Authentication Requirement: Administrator
Function Description: Batch set tags for the specified list of channels

💡 Request Example:

const response = await fetch('/api/channel/batch/tag', {  
  method: 'POST',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  },  
  body: JSON.stringify({  
    ids: [1, 2, 3],  
    tag: "production"  
  })  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "",  
  "data": 3  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "参数错误"  
}

🧾 Field Description:

ids (Array): List of channel IDs to set tags for, required and cannot be empty
tag (String): Name of the tag to be set. Passing null clears the tag.
data (Number): Number of channels successfully tagged

Get Models by Tag¶

Interface Name: Get Models by Tag
HTTP Method: GET
Path: /api/channel/tag/models
Authentication Requirement: Administrator
Function Description: Retrieve the model list that has the highest model count among all channels under the specified tag

💡 Request Example:

const response = await fetch('/api/channel/tag/models?tag=production', {  
  method: 'GET',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  }  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "",  
  "data": "gpt-3.5-turbo,gpt-4,claude-3-sonnet"  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "tag不能为空"  
}

🧾 Field Description:

tag (String): Tag name, required
data (String): Model list of the channel with the most models under this tag, comma-separated

Copy Channel¶

Interface Name: Copy Channel
HTTP Method: POST
Path: /api/channel/copy/:id
Authentication Requirement: Administrator
Function Description: Duplicate an existing channel to create a new one, supporting custom suffixes and balance reset options

💡 Request Example:

const response = await fetch('/api/channel/copy/123?suffix=_备份&reset_balance=true', {  
  method: 'POST',  
  headers: {  
    'Content-Type': 'application/json',  
    'Authorization': 'Bearer your_admin_token',
    'New-Api-User': 'your_user_id'
  }  
});  
const data = await response.json();

✅ Successful Response Example:

{  
  "success": true,  
  "message": "",  
  "data": {  
    "id": 124  
  }  
}

❗ Failed Response Example:

{  
  "success": false,  
  "message": "invalid id"  
}

🧾 Field Description:

id (Number): Channel ID to be copied, passed via URL path
suffix (String): Optional, suffix added to the original name, defaults to "_复制"
reset_balance (Boolean): Optional, whether to reset the balance and used quota to 0, defaults to true
data.id (Number): ID of the newly created channel