Anthropic Messages - 吉吉 Docs

POST https://www.jiji.cc/v1/messages

兼容 Anthropic 官方 Messages 接口，可以完整使用思考、工具调用、图片和 PDF 理解等功能。

基础调用

import os
import anthropic

client = anthropic.Anthropic(
    base_url="https://www.jiji.cc",
    auth_token=os.environ["JIJI_API_KEY"],
)

msg = client.messages.create(
    model="claude-opus-5-5",
    max_tokens=1024,
    system="你是一个简洁的编程助手。",
    messages=[{"role": "user", "content": "用一句话解释什么是闭包。"}],
)
print(msg.content[0].text)

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
  baseURL: "https://www.jiji.cc",
  authToken: process.env.JIJI_API_KEY,
});

const msg = await client.messages.create({
  model: "claude-opus-5-5",
  max_tokens: 1024,
  system: "你是一个简洁的编程助手。",
  messages: [{ role: "user", content: "用一句话解释什么是闭包。" }],
});
console.log(msg.content[0].text);

curl https://www.jiji.cc/v1/messages \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $JIJI_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "claude-opus-5-5",
    "max_tokens": 1024,
    "system": "你是一个简洁的编程助手。",
    "messages": [{"role": "user", "content": "用一句话解释什么是闭包。"}]
  }'

SDK 的 base_url 填 https://www.jiji.cc，不要加 /v1
使用 auth_token（Node.js 为 authToken）传入 Key，SDK 会以 Authorization: Bearer 方式发送

流式输出

with client.messages.stream(
    model="claude-opus-5-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "写一首关于春天的短诗"}],
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)

const stream = client.messages.stream({
  model: "claude-opus-5-5",
  max_tokens: 1024,
  messages: [{ role: "user", content: "写一首关于春天的短诗" }],
});
stream.on("text", (text) => process.stdout.write(text));
await stream.finalMessage();

工具调用

Python

tools = [{
    "name": "get_weather",
    "description": "查询城市的实时天气",
    "input_schema": {
        "type": "object",
        "properties": {"city": {"type": "string"}},
        "required": ["city"],
    },
}]

messages = [{"role": "user", "content": "北京今天天气怎么样？"}]
msg = client.messages.create(
    model="claude-opus-5-5", max_tokens=1024, tools=tools, messages=messages
)

if msg.stop_reason == "tool_use":
    messages.append({"role": "assistant", "content": msg.content})
    results = []
    for block in msg.content:
        if block.type == "tool_use":
            results.append({
                "type": "tool_result",
                "tool_use_id": block.id,
                "content": f"{block.input['city']}：晴，22°C",  # 替换为真实查询
            })
    messages.append({"role": "user", "content": results})
    msg = client.messages.create(
        model="claude-opus-5-5", max_tokens=1024, tools=tools, messages=messages
    )

print(msg.content[0].text)

图片理解

Python

import base64

with open("photo.png", "rb") as f:
    b64 = base64.b64encode(f.read()).decode()

msg = client.messages.create(
    model="claude-opus-5-5",
    max_tokens=1024,
    messages=[{
        "role": "user",
        "content": [
            {"type": "image", "source": {"type": "base64", "media_type": "image/png", "data": b64}},
            {"type": "text", "text": "这张图片里有什么？"},
        ],
    }],
)
print(msg.content[0].text)

PDF 文件使用 "type": "document"、"media_type": "application/pdf"，写法相同。

常用参数

参数	说明
`model`	模型 ID，以模型广场为准
`max_tokens`	最多生成的 Token 数，必填
`system`	系统提示词
`messages`	对话消息列表
`tools`	可供模型调用的工具列表
`stream`	是否流式返回