返回首页

API 参考

答安 API 完全兼容 OpenAI 接口格式，所有请求发送至统一 Base URL：

Base URL

https://model.token618.com/api/open-apis/v1

认证方式

Authorization: Bearer YOUR_API_KEY

接口概览

方法路径说明
POST/chat/completions对话补全（核心接口）
POST/images/generations图像生成
POST/embeddings文本向量化
GET/models获取可用模型列表

Chat Completions

最常用的对话补全接口，支持流式（SSE）和非流式响应，兼容 OpenAI SDK 直接调用。

POST /chat/completions

请求参数

参数类型必填说明
modelstring必填模型 ID（如 GLM-5.1、DeepSeek-V3.2）
messagesarray必填消息列表，每条包含 role（system / user / assistant / tool）和 content
streamboolean选填是否使用流式输出（默认 false）
temperaturenumber选填采样温度，0-2 之间（默认 1）。值越高输出越随机
max_tokensinteger选填最大生成 Token 数。不设则使用模型默认值
top_pnumber选填核采样概率阈值（默认 1），与 temperature 二选一
stopstring / array选填停止生成的标记，最多 4 个
toolsarray选填可用的函数工具列表（见函数调用章节）
tool_choicestring / object选填控制函数调用行为：auto、none、或指定函数名
response_formatobject选填指定输出格式，如 { "type": "json_object" }
stream_optionsobject选填流式选项，设为 { "include_usage": true } 可在最后获取用量

非流式请求示例

Request
curl https://model.token618.com/api/open-apis/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "GLM-5.1",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant."},
      {"role": "user", "content": "Hello!"}
    ],
    "stream": false
  }'
Response
{
  "id": "chatcmpl-xxx",
  "object": "chat.completion",
  "model": "GLM-5.1",
  "choices": [{
    "index": 0,
    "message": {
      "role": "assistant",
      "content": "Hello! How can I help you today?"
    },
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 20,
    "completion_tokens": 8,
    "total_tokens": 28
  }
}

流式输出（Streaming / SSE）

设置 stream: true 后， 接口返回 Server-Sent Events (SSE) 格式的增量响应。每个 chunk 包含一小段生成的文本， 最后以 [DONE] 标记结束。

流式请求示例

Request
curl https://model.token618.com/api/open-apis/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "GLM-5.1",
    "messages": [{"role": "user", "content": "Hello!"}],
    "stream": true
  }'
Response
data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"role":"assistant","content":""},"finish_reason":null}]}

data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"Hello"},"finish_reason":null}]}

data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":"!"},"finish_reason":null}]}

data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":" How"},"finish_reason":null}]}

data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":" can I help"},"finish_reason":null}]}

data: {"id":"chatcmpl-xxx","object":"chat.completion.chunk","choices":[{"index":0,"delta":{"content":" you today?"},"finish_reason":"stop"}]}

data: [DONE]

流式用量统计

默认情况下流式响应不包含 usage 字段。如需获取 Token 用量，添加参数：

stream_options 参数
{
  "model": "GLM-5.1",
  "messages": [{"role": "user", "content": "Hello!"}],
  "stream": true,
  "stream_options": { "include_usage": true }
}

设置后，在 [DONE] 之前的最后一个 chunk 将包含完整的 usage 统计。

SDK 调用示例

Python (OpenAI SDK)
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://model.token618.com/api/open-apis/v1"
)

stream = client.chat.completions.create(
    model="GLM-5.1",
    messages=[{"role": "user", "content": "Hello!"}],
    stream=True,
    stream_options={"include_usage": True},
)

for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")
    if chunk.usage:
        print(f"\nUsage: {chunk.usage}")
Node.js (OpenAI SDK)
import OpenAI from "daan";

const client = new OpenAI({
  apiKey: "YOUR_API_KEY",
  baseURL: "https://model.token618.com/api/open-apis/v1",
});

const stream = await client.chat.completions.create({
  model: "GLM-5.1",
  messages: [{ role: "user", content: "Hello!" }],
  stream: true,
  stream_options: { include_usage: true },
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? "");
  if (chunk.usage) {
    console.log("\nUsage:", chunk.usage);
  }
}

函数调用（Function Calling / Tool Use）

支持通过 tools 参数让模型调用自定义函数。 模型会在需要时返回函数调用请求，你的代码执行后将结果传回模型继续对话。

请求示例

Python (Function Calling)
from openai import OpenAI
import json

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://model.token618.com/api/open-apis/v1"
)

# Step 1: 发送带 tools 的请求
response = client.chat.completions.create(
    model="GLM-5.1",
    messages=[{"role": "user", "content": "北京今天天气怎么样？"}],
    tools=[{
        "type": "function",
        "function": {
            "name": "get_weather",
            "description": "获取指定城市的天气信息",
            "parameters": {
                "type": "object",
                "properties": {
                    "city": {
                        "type": "string",
                        "description": "城市名称"
                    }
                },
                "required": ["city"]
            }
        }
    }],
    tool_choice="auto"
)

message = response.choices[0].message

# Step 2: 检查模型是否要调用函数
if message.tool_calls:
    tool_call = message.tool_calls[0]
    args = json.loads(tool_call.function.arguments)

    # Step 3: 执行你的函数逻辑
    weather_result = get_weather(args["city"])

    # Step 4: 将结果返回给模型
    response2 = client.chat.completions.create(
        model="GLM-5.1",
        messages=[
            {"role": "user", "content": "北京今天天气怎么样？"},
            message,
            {
                "role": "tool",
                "tool_call_id": tool_call.id,
                "content": json.dumps(weather_result, ensure_ascii=False)
            },
        ]
    )
    print(response2.choices[0].message.content)
Response
// 模型返回 tool_calls 时的响应格式
{
  "choices": [{
    "message": {
      "role": "assistant",
      "content": null,
      "tool_calls": [{
        "id": "call_abc123",
        "type": "function",
        "function": {
          "name": "get_weather",
          "arguments": "{\"city\":\"北京\"}"
        }
      }]
    },
    "finish_reason": "tool_calls"
  }]
}

finish_reason 取值说明

响应中 finish_reason 字段表示生成结束的原因：

值说明
stop模型正常完成输出，或遇到了 stop 序列
length输出达到 max_tokens 限制而被截断
tool_calls模型决定调用函数
content_filter内容被安全过滤器拦截
error生成过程中发生错误（流式场景下 HTTP 状态码仍为 200）

Images

图像生成接口，支持 DALL-E 等模型。

POST /images/generations
Request
curl https://model.token618.com/api/open-apis/v1/images/generations \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "z-image-turbo",
    "prompt": "A cute cat wearing a hat",
    "n": 1,
    "size": "1024x1024"
  }'

Embeddings

文本向量化接口，支持多种 Embedding 模型。

POST /embeddings
Request
curl https://model.token618.com/api/open-apis/v1/embeddings \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "text-embedding-3-small",
    "input": "Hello world"
  }'

Models

获取当前账户可用的模型列表。

GET /models
Request
curl https://model.token618.com/api/open-apis/v1/models \
  -H "Authorization: Bearer YOUR_API_KEY"
Response
{
  "object": "list",
  "data": [
    {
      "id": "GLM-5.1",
      "object": "model",
      "owned_by": "daan"
    },
    {
      "id": "deepseek-chat",
      "object": "model",
      "owned_by": "deepseek"
    }
  ]
}

错误处理

答安 API 采用两级错误返回机制，确保不同场景下错误信息可被正确捕获。

非流式请求 / 流式请求开始前

返回标准 HTTP 错误状态码和 JSON 错误体，可正常通过 HTTP 状态码判断错误类型。

Response
{
  "error": {
    "code": 401,
    "message": "Invalid API key provided",
    "metadata": {}
  }
}

流式请求进行中（Mid-stream Error）

如果错误发生在流式传输过程中，HTTP 状态码仍为 200，但 SSE 事件中会包含错误信息， 同时 finish_reason 为 "error"。

Response
data: {"error":{"code":502,"message":"Provider error"},"choices":[{"index":0,"delta":{"content":null},"finish_reason":"error"}]}

data: [DONE]

错误码一览

状态码类型说明
400参数错误请求参数格式错误或缺少必填参数
401认证失败API Key 无效、未提供或已过期
402余额不足账户余额不足以支付本次请求
403权限拒绝无权访问该模型或资源
408请求超时请求处理超时，建议稍后重试
429频率限制请求频率超限，请降低并发或联系客服提升限额
500服务错误服务端内部错误，答安会自动重试或切换渠道
502上游错误上游供应商返回错误，答安会自动尝试备用渠道
503暂时不可用模型暂时不可用，通常几分钟后恢复

重试建议

  • 遇到 429/502/503 时建议使用指数退避重试（Exponential Backoff）
  • 401/402/403 属于业务错误，重试不会生效，请检查 API Key 和余额
  • 答安网关已内置故障转移机制，500/502 会自动尝试备用渠道

下一步

了解如何配置 CLI 工具请查看 CLI 工具配置， 常见问题请查看 常见问题。

下一页概述