万界方舟 Open API 接口文档

接入流程说明

访问万界方舟-模型平台，通过手机号登录，在“模型广场”中，点击模型名称进行模型复制。
获取密钥：登录平台 - 在右上角头像 - “个人中心” - APIKey查看。
模型说明：开放的模型会在“模型广场”中自动分配，其他模型服务需要和商业运营进行授权开通。
所有公开开放的模型完全兼容 openai 接口，私有订制模型见万界方舟外部接口文档。

接口基础信息

Base Url： https://maas-openapi.wanjiedata.com/api
Api Key：在个人中心 - APIKey栏目授权 APIKEY 获取。

获取模型列表

获取已经授权的模型列表。

shellpythonjavascriptcsharp

shell


export API_KEY="<你的 API KEY>"
curl https://maas-openapi.wanjiedata.com/api/v1/models \
  -H "Authorization: Bearer $API_KEY"

python

import requests
API_KEY = "<你的 API KEY>"
url = "https://maas-openapi.wanjiedata.com/api/v1/models"
headers = {
    "Authorization": f"Bearer {API_KEY}"
}
response = requests.get(url, headers=headers)
print(response.status_code)
print(response.json())

javascript

import axios from 'axios';

const apiKey = '<你的 API KEY>';

async function getModels() {
  try {
    const response = await axios.get('https://maas-openapi.wanjiedata.com/api/v1/models', {
      headers: { Authorization: `Bearer ${apiKey}` }
    });
    console.log('模型列表:', response.data);
  } catch (error) {
    console.error('请求失败:', error.response ? error.response.data : error.message);
  }
}

getModels();

csharp

using System;
using System.Net.Http;
using System.Threading.Tasks;

class Program
{
   static async Task Main()
   {
       var client = new HttpClient();
       client.DefaultRequestHeaders.Add("Authorization", "Bearer <你的 API KEY>");
       var response = await client.GetStringAsync("https://maas-openapi.wanjiedata.com/api/v1/models");
       Console.WriteLine(response);
   }
}

响应示例

shellpythonjavascriptcsharp

shell

[
    {
        "id": "deepseek-v3",
        "object": "model",
        "created": 1740362631,
        "owned_by": "wanjie"
    },
    {
        "id": "deepseek-reasoner",
        "object": "model",
        "created": 1739017876,
        "owned_by": "wanjie"
    },
    },
    },
]

python

{
  'data': [
    {'id': 'qwen3-32b', 'object': 'model', 'created': 1760084423, 'owned_by': 'organization'},
    {'id': 'claude-3.7-sonnet-20250219', 'object': 'model', 'created': 1760083578, 'owned_by': 'organization'},
    {'id': 'Flux.1-Schnell', 'object': 'model', 'created': 1760083578, 'owned_by': 'organization'},
    {'id': 'deepseek-v3', 'object': 'model', 'created': 1760083578, 'owned_by': 'organization'}
  ],
  'object': 'list'
}

javascript

模型列表: {
  data: [
    {id: 'qwen3-32b', object: 'model', created: 1760084423, owned_by: 'organization'},
    {id: 'claude-3.7-sonnet-20250219', object: 'model', created: 1760083578, owned_by: 'organization'},
    {id: 'Flux.1-Schnell', object: 'model', created: 1760083578, owned_by: 'organization'},
    {id: 'deepseek-v3', object: 'model', created: 1760083578, owned_by: 'organization'}
  ],
  object: 'list'
}

csharp

d":1760083578,"owned_by":"organization"},
{"id":"gemini-2.5-flash-lite","object":"model","created":1760083578,"owned_by":"organization"},
{"id":"gemini-2.5-flash","object":"model","created":1760083578,"owned_by":"organization"},
{"id":"gemini-2.5-pro","object":"model","created":1760083578,"owned_by":"organization"},
{"id":"claude-sonnet-4-20250514","object":"model","created":1760083578,"owned_by":"organization"},
{"id":"GPT-4o","object":"model","created":1760083578,"owned_by":"organization"},
{"id":"GPT-4.1","object":"model","created":1760083578,"owned_by":"organization"},
{"id":"deepseek-reasoner","object":"model","created":1760083578,"owned_by":"organization"},
{"id":"qwen3-next","object":"model","created":1760083578,"owned_by":"organization"},
{"id":"qwen3-coder","object":"model","created":1760083578,"owned_by":"organization"},
{"id":"kimi-k2","object":"model","created":1760083578,"owned_by":"organization"},
{"id":"qwen3-235b","object":"model","created":1760083578,"owned_by":"organization"},
{"id":"CosyVoice","object":"model","created":1760083578,"owned_by":"organization"},
{"id":"Flux.1-Schnell","object":"model","created":1760083578,"owned_by":"organization"},
{"id":"deepseek-v3","object":"model","created":1760083578,"owned_by":"organization"}],"object":"list"}

聊天对话接口

shell

POST /v1/chat/completions

非流式请求示例

shellpythonjavascriptcsharp

shell

# 调用聊天对话请求示例
export API_KEY="<你的 API KEY>"
export MODEL="<你的授权模型名称>"
# 授权模型名称：在模型广场列表中模型名称复制名称获取。
curl --location --request POST 'https://maas-openapi.wanjiedata.com/api/v1/chat/completions' \
--header "Authorization: Bearer $API_KEY" \
--header 'Content-Type: application/json' \
--data-raw '{
    "model": "'$MODEL'",
    "messages": [{"role": "user", "content": "Hello!"}],
    "stream": false
}'

python

import requests
url = "https://maas-openapi.wanjiedata.com/api/v1/chat/completions"
api_key = "<你的 API KEY>"
headers = {
    "Authorization": f"Bearer {api_key}",
    "Content-Type": "application/json"
}
data = {
    "model": "<你的授权模型名称>",
    "messages": [
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello!"}
    ]
}
response = requests.post(url, headers=headers, json=data)
response_json = response.json()
assistant_message = response_json['choices'][0]['message']['content']
print(assistant_message)

javascript

import axios from 'axios';

const apiKey = '<你的 API KEY>';
const model = '<你的授权模型名称>'; 
const url = 'https://maas-openapi.wanjiedata.com/api/v1/chat/completions';

async function testConnection() {
  try {
    const response = await axios.post(url, {
      model,
      messages: [
        { role: 'user', content: 'Hello!' }
      ]
    }, {
      headers: {
        Authorization: `Bearer ${apiKey}`,
        'Content-Type': 'application/json'
      }
    });
    console.log(response.data.choices[0].message.content);
    console.log('API 连接成功');
  } catch (error) {
    console.error('API 连接失败:', error.response?.data || error.message);
  }
}

testConnection();

csharp

using System.Net.Http;
using System.Text;
using System.Threading.Tasks;

class P {
    static async Task Main() {
        var c = new HttpClient();
        c.DefaultRequestHeaders.Add("Authorization", "Bearer <API_KEY>");
        var d = new StringContent(@"{
            ""model"":""<MODEL>"",
            ""messages"":[{""role"":""user"",""content"":""Hello!""}],
            ""stream"":false
        }", Encoding.UTF8, "application/json");
        var r = await c.PostAsync("https://maas-openapi.wanjiedata.com/api/v1/chat/completions", d);
        System.Console.WriteLine(await r.Content.ReadAsStringAsync());
    }
}

非流式响应示例

shellpythonjavascriptcsharp

shell

{
    "id": "chatcmpl-xxxxxx",
    "object": "chat.completion",
    "created": 1760336462,
    "model": "deepseek-reasoner",
    "choices": [
        {
            "index": 0,
            "message": {
                "role": "assistant",
                "content": "Hello! 😊 How can I assist you today?"
            },
            "finish_reason": "stop"
        }
    ],
    "usage": {
        "prompt_tokens": 8,
        "completion_tokens": 228,
        "total_tokens": 236
    }
}

python

Hello! 😊 How can I assist you today?

javascript

Hello! 😊 How can I assist you today? Whether you have a question, need help with something, or just want to chat, I'm here for you!
API 连接成功

csharp

{
  "id":"chatcmpl-187830d79894458aa702fd592b109d4b",
"object":"chat.completion","created":1762412451,
"model":"deepseek-reasoner","choices":[{"index":0,"message":{"role":"assistant",
"content":"\nHi there! 😊  \nHow are you doing today? What can I help you with? 😊",
"tool_calls":[]},"finish_reason":"stop"}],
"usage":{"prompt_tokens":5,"completion_tokens":186,"total_tokens":191}
}

流式请求示例

shellpythonjavascriptcsharp

shell

# 调用聊天对话请求示例
export API_KEY="<你的 API KEY>"
export MODEL="<你的授权模型名称>"
# 授权模型名称：在授权模型列表中模型名称复制名称获取。
curl --location --request POST 'https://maas-openapi.wanjiedata.com/api/v1/chat/completions' \
--header "Authorization: Bearer $API_KEY" \
--header 'Content-Type: application/json' \
--data-raw '{
    "model": "'$MODEL'",
    "messages": [{"role": "user", "content": "请用一句话介绍自己"}],
    "stream": true
}'

python

import requests

API_KEY = "<你的 API KEY>"
MODEL = "<你的授权模型名称>"

url = 'https://maas-openapi.wanjiedata.com/api/v1/chat/completions'
headers = {
    'Authorization': f'Bearer {API_KEY}',
    'Content-Type': 'application/json'
}
data = {
    "model": MODEL,
    "messages": [{"role": "user", "content": "请用一句话介绍自己"}],
    "stream": True
}

response = requests.post(url, headers=headers, json=data)
print(response.text)

javascript

fetch('https://maas-openapi.wanjiedata.com/api/v1/chat/completions', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer <你的 API KEY>',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    model: '<你的授权模型名称>',
    messages: [{ role: 'user', content: '请用一句话介绍自己' }],
    stream: true
  })
})
.then(res => res.text())
.then(console.log);

csharp

using System.Net.Http;
using System.Text;
using System.Threading.Tasks;

class P {
    static async Task Main() {
        var c = new HttpClient();
        c.DefaultRequestHeaders.Add("Authorization", "Bearer <API_KEY>");
        var d = new StringContent(@"{
            ""model"":""<MODEL>"",
            ""messages"":[{""role"":""user"",""content"":""请用一句话介绍自己""}],
            ""stream"":true
        }", Encoding.UTF8, "application/json");
        var r = await c.PostAsync("https://maas-openapi.wanjiedata.com/api/v1/chat/completions", d);
        System.Console.WriteLine(await r.Content.ReadAsStringAsync());
    }
}

流式响应示例

shellpythonjavascriptcsharp

shell

{
    "id": "chatcmpl-5cb3566f274045d29a948339258bf4fe",
    "object": "chat.completion.chunk",
    "created": 1762307854,
    "model": "deepseek-reasoner",
    "system_fingerprint": "fp_931335",
    "choices": [
        {
            "index": 0,
            "delta": {
                "role": "assistant",
                "content": null,
                "reasoning_content": "",
                ...
                ...
                "refusal": null,
                "tool_calls": null
            },
            "logprobs": null,
            "finish_reason": null
        }
    ],
    "usage": null
}
[DONE
]

python

id: 0
data: {"id":"chatcmpl-d9a6e351096347aba63d25ecb858c1ab","object":"chat.completion.chunk","created":1762307411,"model":"deepseek-reasoner","system_fingerprint":"fp_102727","choices":[{"index":0,"delta":{"role":"assistant","content":null,"reasoning_content":"","refusal":null,"tool_calls":null},"logprobs":null,"finish_reason":null}],"usage":null}
id: 1
data: {"id":"chatcmpl-d9a6e351096347aba63d25ecb858c1ab","object":"chat.completion.chunk","created":1762307411,"model":"deepseek-reasoner","system_fingerprint":"fp_082062","choices":[{"index":0,"delta":{"role":"assistant","content":null,"reasoning_content":"\n","refusal":null,"tool_calls":null},"logprobs":null,"finish_reason":null}],"usage":null}
...
id: 214
data: [DONE]

javascript

id: 0
data: {"id":"chatcmpl-21162e8a7db647c4b6322a08a8642b6a","object":"chat.completion.chunk","created":1762307630,"model":"deepseek-reasoner","system_fingerprint":"fp_305980","choices":[{"index":0,"delta":{"role":"assistant","content":null,"reasoning_content":"","refusal":null,"tool_calls":null},"logprobs":null,"finish_reason":null}],"usage":null}
id: 1
data: {"id":"chatcmpl-21162e8a7db647c4b6322a08a8642b6a","object":"chat.completion.chunk","created":1762307630,"model":"deepseek-reasoner","system_fingerprint":"fp_628390","choices":[{"index":0,"delta":{"role":"assistant","content":null,"reasoning_content":"\n","refusal":null,"tool_calls":null},"logprobs":null,"finish_reason":null}],"usage":null}
...
id: 214
data: [DONE]

csharp

id: 0
data: {"id":"chatcmpl-84478138cdd54db8a6fe08ca6303d988","object":"chat.completion.chunk","created":1762413495,"model":"deepseek-reasoner","system_fingerprint":"fp_397373","choices":[{"index":0,"delta":{"role":"assistant","content":null,"reasoning_content":"","refusal":null,"tool_calls":null},"logprobs":null,"finish_reason":null}],"usage":null}

...

id: 213
data: {"id":"chatcmpl-84478138cdd54db8a6fe08ca6303d988","object":"chat.completion.chunk","created":1762413495,"model":"deepseek-reasoner","system_fingerprint":"fp_065818","choices":[],"usage":{"prompt_tokens":8,"completion_tokens":214,"total_tokens":222,"prompt_tokens_details":null,"completion_tokens_details":null}}

id: 214
data: [DONE]

聊天对话接口参数说明见 OpenAI API Chat completions

文字转语音接口

支持的模型：CosyVoice

shell

POST /v1/audio/speech

shellpythonjavascriptcsharp

shell

export API_KEY="<你的 API KEY>"
# 授权文字转语音模型名称：在授权模型列表中模型名称复制名称获取。
export MODEL="<你的授权文字转语音模型名称>"
curl --location --request POST 'https://maas-openapi.wanjiedata.com/api/v1/audio/speech' \
--header "Authorization: Bearer $API_KEY" \
--header 'Content-Type: application/json' \
--data-raw '{
  "input": "您好，我是文字转语音助手，一个由万界数据训练的语音合成模型，你可以尝试输入一段文字，将帮您把这段文字转换为语音输出。",
  "model": "'$MODEL'",
  "voice": "普通话",
  "responseFormat": "mp3",
  "speed": 1.0
}' \
--output output.mp3

python

import requests
API_KEY = "<你的 API KEY>"
MODEL = "<你的授权文字转语音模型名称>"
url = "https://maas-openapi.wanjiedata.com/api/v1/audio/speech"
headers = {
    "Authorization": f"Bearer {API_KEY}",
    "Content-Type": "application/json"
}
payload = {
    "input": "您好，我是文字转语音助手，一个由万界数据训练的语音合成模型，你可以尝试输入一段文字，将帮您把这段文字转换为语音输出。",
    "model": MODEL,
    "voice": "普通话",
    "responseFormat": "mp3",
    "speed": 1.0
}
response = requests.post(url, headers=headers, json=payload)
if response.status_code == 200:
    with open("output.mp3", "wb") as f:
        f.write(response.content)
    print("音频已保存为 output.mp3")
else:
    print("请求失败:", response.status_code)
    print(response.text)

javascript

import axios from 'axios';
import fs from 'fs';

const apiKey = '<你的 API KEY>'; 
const model = '<你的授权文字转语音模型名称>';       
const url = 'https://maas-openapi.wanjiedata.com/api/v1/audio/speech';
const inputText = '你好，我是文字转语音助手！';

async function textToSpeech() {
  try {
    const response = await axios.post(url, {
      input: inputText,
      model: model,
      voice: '普通话',
      responseFormat: 'mp3',
      speed: 1.0
    }, {
      headers: {
        Authorization: `Bearer ${apiKey}`,
        'Content-Type': 'application/json'
      },
      responseType: 'arraybuffer'
    });

    fs.writeFileSync('output.mp3', response.data);
    console.log('MP3文件已保存为 output.mp3');
  } catch (error) {
    console.error('请求失败:', error.response?.data || error.message);
  }
}

textToSpeech();

csharp

using System.Net.Http;
using System.Text;
using System.Threading.Tasks;
using System.IO;

class P {
    static async Task Main() {
        var c = new HttpClient();
        c.DefaultRequestHeaders.Add("Authorization", "Bearer <API_KEY>");
        var d = new StringContent(@"{
            ""input"":""您好，我是文字转语音助手，一个由万界数据训练的语音合成模型，你可以尝试输入一段文字，将帮您把这段文字转换为语音输出。"", 
            ""model"":""<MODEL>"",
            ""voice"":""普通话"",
            ""responseFormat"":""mp3"",
            ""speed"":1.0
        }", Encoding.UTF8, "application/json");
        var r = await c.PostAsync("https://maas-openapi.wanjiedata.com/api/v1/audio/speech", d);
        var bytes = await r.Content.ReadAsByteArrayAsync();
        await File.WriteAllBytesAsync("output.mp3", bytes);
    }
}

响应示例

shellpythonjavascriptcsharp

shell

# 文件 output.mp3 已保存为当前目录

python

音频已保存为 output.mp3

javascript

MP3文件已保存为 output.mp3

csharp


    Directory: D:\C#\work\ModelProject

Mode                 LastWriteTime         Length Name
----                 -------------         ------ ----
-a---           2025/11/6    15:23         176910 output.mp3

文转语音接口参数说明见 OpenAI API Create speech

文生图接口

支持的模型：Flux.1-Schnell（该模型提示词只支持英文）

shell

POST /v1/images/generations

shellpythonjavascriptcsharp

shell

export API_KEY="<你的 API KEY>"
export MODEL="<你的授权文生图模型名称>"
curl --location --request POST 'https://maas-openapi.wanjiedata.com/api/v1/images/generations' \
--header "Authorization: Bearer $API_KEY"  \
--header 'Content-Type: application/json' \
--data-raw '{
  "model": "'$MODEL'",
  "prompt": "graphic of a Harley Davidson bike",
  "n": 1,
  "size": "1024x1024",
  "responseFormat": "url",
  "quality": "standard",
  "style": "vivid"
}'

python

import requests
API_KEY = "<你的 API KEY>"
MODEL = "<你的授权文生图模型名称>"
url = "https://maas-openapi.wanjiedata.com/api/v1/images/generations"
headers = {
    "Authorization": f"Bearer {API_KEY}",
    "Content-Type": "application/json"
}
payload = {
    "model": MODEL,
    "prompt": "graphic of a Harley Davidson bike",
    "n": 1,
    "size": "1024x1024",
    "responseFormat": "url",
    "quality": "standard",
    "style": "vivid"
}
response = requests.post(url, headers=headers, json=payload)
if response.status_code == 200:
    result = response.json()
    print(result) 
else:
    print("请求失败:", response.status_code)
    print(response.text)

javascript

import axios from 'axios';

const apiKey = '<你的 API KEY>'; 
const model = '<你的授权文生图模型名称>'; 
const url = 'https://maas-openapi.wanjiedata.com/api/v1/images/generations';

async function generateImage() {
  const headers = {
    Authorization: `Bearer ${apiKey}`,
    'Content-Type': 'application/json'
  };
  const data = {
    model,
    prompt: 'graphic of a Harley Davidson bike',
    n: 1,
    size: '1024x1024',
    responseFormat: 'url'
  };

  try {
    const res = await axios.post(url, data, { headers });
    console.log(res.data.data[0].url); 
  } catch (err) {
    console.error('请求失败:', err.message);
  }
}

generateImage();

csharp

using System.Net.Http;
using System.Text;
using System.Threading.Tasks;

class P {
    static async Task Main() {
        var c = new HttpClient();
        c.DefaultRequestHeaders.Add("Authorization", "Bearer <API_KEY>");
        var d = new StringContent(@"{
            ""model"":""<MODEL>"",
            ""prompt"":""graphic of a Harley Davidson bike"",
            ""n"":1,
            ""size"":""1024x1024"",
            ""responseFormat"":""url"",
            ""quality"":""standard"",
            ""style"":""vivid""
        }", Encoding.UTF8, "application/json");
        var r = await c.PostAsync("https://maas-openapi.wanjiedata.com/api/v1/images/generations", d);
        System.Console.WriteLine(await r.Content.ReadAsStringAsync());
    }
}

响应示例

shellpythonjavascriptcsharp

shell

{
    "created": 1762221608,
    "data": [
        {
            "url": "https://rgw.wanjiedata.com/maas-public-bucket/2025/11/04/c3fe6bdd3a601f90c4d1e33e961e064f.jpg"
        }
    ],
    "usage": {
        "input_tokens_details": {}
    }
}

python

{
    'created': 1762220795, 
    'data': [{'url': 'https://rgw.wanjiedata.com/maas-public-bucket/2025/11/04/a920b1331c7c5676624e38cbc10d8b43.jpg'}],
    'usage': {'input_tokens_details': {}
    }
    }

javascript

https://rgw.wanjiedata.com/maas-public-bucket/2025/11/04/0d4e5bc27fe7b76a1544a36d75f4c1da.jpg

csharp

{"created":1762414008,"data":[{"url":"https://rgw.wanjiedata.com/maas-public-bucket/2025/11/06/6c6eaaa374eb4be4a629b3adff530412.jpg"}],"usage":{"input_tokens_details":{}}}

对话助手

翻译工具

辅助写作

开发工具&平台

智能搜索

灵感创作

对话助手

翻译工具

辅助写作

开发工具&平台

智能搜索

灵感创作

万界方舟 Open API 接口文档

接入流程说明

接口基础信息

获取模型列表

响应示例

聊天对话接口

聊天对话接口参数说明见 OpenAI API Chat completions

文字转语音接口

文转语音接口参数说明见 OpenAI API Create speech

文生图接口

文生图接口参数说明见 OpenAI API Create image

万界方舟 Open API 接口文档 ​

接入流程说明 ​

接口基础信息 ​

获取模型列表 ​

响应示例 ​

聊天对话接口 ​

聊天对话接口参数说明见 OpenAI API Chat completions ​

文字转语音接口 ​

文转语音接口参数说明见 OpenAI API Create speech ​

文生图接口 ​

文生图接口参数说明见 OpenAI API Create image ​

万界方舟 Open API 接口文档

接入流程说明

接口基础信息

获取模型列表

响应示例

聊天对话接口

聊天对话接口参数说明见 OpenAI API Chat completions

文字转语音接口

文转语音接口参数说明见 OpenAI API Create speech

文生图接口

文生图接口参数说明见 OpenAI API Create image