搭配 LiteLLM 的 CopilotKit SDK

透過 LiteLLM Proxy 使用 CopilotKit SDK 搭配任何 LLM 提供者。

注意： CopilotKit SDK 與 LiteLLM Proxy 的整合可搭配 LiteLLM v1.81.7-nightly 或更高版本使用。

快速開始

1. 將模型新增至設定

config.yaml

model_list:
  - model_name: claude-sonnet-4-5
    litellm_params:
      model: "anthropic/claude-sonnet-4-5-20250514-v1:0"
      api_key: "os.environ/ANTHROPIC_API_KEY"

2. 啟動 LiteLLM Proxy

litellm --config config.yaml

3. 使用 CopilotKit SDK

import OpenAI from "openai";
import {
  CopilotRuntime,
  OpenAIAdapter,
  copilotRuntimeNextJSAppRouterEndpoint,
} from "@copilotkit/runtime";
import { NextRequest } from "next/server";

const model = "claude-sonnet-4-5";

const openai = new OpenAI({
  apiKey: process.env.OPENAI_API_KEY || "sk-12345",
  baseURL: process.env.OPENAI_BASE_URL || "http://localhost:4000/v1",
});

const serviceAdapter = new OpenAIAdapter({ openai, model });
const runtime = new CopilotRuntime();

export const POST = async (req: NextRequest) => {
  const { handleRequest } = copilotRuntimeNextJSAppRouterEndpoint({
    runtime,
    serviceAdapter,
    endpoint: "/api/copilotkit",
  });
  return handleRequest(req);
};

4. 測試

curl -X POST http://localhost:3000/api/copilotkit \
  -H "Content-Type: application/json" \
  -d '{
    "method": "agent/run",
    "params": {
        "agentId": "default"
    },
        "runId": "your_run_id",
        "threadId": "your_thread_id",
        "runId": ""your_run_id"",
        "tools": [],
        "context": [],
        "forwardedProps": {},
        "state": {},
        "messages": [
            {
                "id": "166e573e-f7c6-4c0f-8685-04dbefec18be",
                "content": "Hi",
                "role": "user"
            }
        ]
    }
}'

環境變數

變數	值	說明
OPENAI_API_KEY	sk-12345	您的 LiteLLM API 金鑰
OPENAI_BASE_URL	http://localhost:4000/v1	LiteLLM proxy URL

相關資源

• CopilotKit 文件
• LiteLLM Proxy 快速開始