OpenAI-Compatible Claude API — Use Claude with OpenAI SDK

TL;DR

Why OpenAI Compatibility Matters

Most AI tools, frameworks, and IDEs were built around the OpenAI API format (/v1/chat/completions). If you have existing code using the OpenAI SDK, migrating to Claude normally means rewriting your API calls to use Anthropic's Messages format.

LLMsRelay's OpenAI-compatible endpoint eliminates this problem. You get Claude's superior coding and reasoning capabilities while keeping your existing OpenAI-format code.

Quick Migration: OpenAI → Claude

from openai import OpenAI

client = OpenAI(api_key="sk-openai-key")

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Write a Python function"}]
)
print(response.choices[0].message.content)

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_LLMSRELAY_KEY",
    base_url="https://api.llmsrelay.com/v1"  # Only change!
)

response = client.chat.completions.create(
    model="claude-sonnet-4.6",  # Use Claude model ID
    messages=[{"role": "user", "content": "Write a Python function"}]
)
print(response.choices[0].message.content)

That's it — two lines changed: base URL and model name.

TypeScript / Node.js Example

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_LLMSRELAY_KEY",
  baseURL: "https://api.llmsrelay.com/v1",
});

const response = await client.chat.completions.create({
  model: "claude-sonnet-4.6",
  messages: [{ role: "user", content: "Explain async/await" }],
});

console.log(response.choices[0].message.content);

Supported OpenAI Endpoints

• POST /v1/chat/completions — Chat completions (streaming & non-streaming)
• GET /v1/models — List available Claude models

Framework & Tool Compatibility

The OpenAI-compatible endpoint works with any tool that supports custom OpenAI base URLs:

• LangChain — use ChatOpenAI with custom base URL
• LlamaIndex — configure OpenAI-compatible LLM
• Cursor IDE — set Override OpenAI Base URL
• Continue (VS Code) — set base URL in config
• AutoGen — use OpenAI-compatible config
• CrewAI — configure as OpenAI provider

Model Mapping

When migrating from OpenAI, here's the recommended Claude equivalent:

Native Anthropic API Also Available

If you prefer the native Anthropic Messages API format, LLMsRelay supports that too:

import anthropic

client = anthropic.Anthropic(
    api_key="YOUR_LLMSRELAY_KEY",
    base_url="https://api.llmsrelay.com"
)

message = client.messages.create(
    model="claude-sonnet-4.6",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello!"}]
)

Use whichever format fits your project best — both hit the same Claude models at the same pricing.