Claude Code Proxy: Switch your client in three lines

Prerequisites: Install Dependencies

Before making requests, ensure your environment has the necessary tools. For command-line testing, you need curl. For Python projects, install the official OpenAI SDK via pip install openai. For Node.js, run npm install openai. These libraries handle the OpenAI-compatible protocol automatically, allowing you to swap the endpoint without altering your application logic.

Configure Base URL and Key

Point your client to our specialized claude code proxy endpoint. Replace the standard OpenAI base URL with our host and provide your unique API key. This single configuration change directs all traffic to our uncensored model server while maintaining full compatibility with standard SDK methods.

Copycurl https://api.claudecodeproxyapi.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "Write a blunt product review of a cheap VPN."}]
  }'

Set BASE_URL to https://api.claudecodeproxyapi.com/v1 and pass your key in the Authorization header. This setup works for any client that supports OpenAI-compatible endpoints.

Test with Chat Completions

Verify connectivity with a simple text request. Send a POST request to /v1/chat/completions with the uncensored model ID. This confirms your key is valid and the proxy is routing requests correctly. The response will contain the model's direct answer without standard content filters for lawful adult content.

Copyfrom openai import OpenAI

client = OpenAI(base_url="https://api.claudecodeproxyapi.com/v1", api_key="YOUR_KEY")

resp = client.chat.completions.create(
    model="uncensored",
    messages=[{"role": "user", "content": "Summarise this thread without softening it."}],
)
print(resp.choices[0].message.content)

Use the openai Python client to send a message. The code mirrors standard OpenAI usage, ensuring zero friction when switching from other providers.

Enable Streaming Responses

For lower latency and real-time token display, enable streaming. Set stream=True in your client configuration. The API returns Server-Sent Events (SSE) as tokens are generated. This is ideal for CLI interfaces or applications where immediate feedback improves user experience.

Copyimport OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.claudecodeproxyapi.com/v1", apiKey: process.env.API_KEY });

const resp = await client.chat.completions.create({
  model: "uncensored",
  messages: [{ role: "user", content: "Draft a villain monologue for my game." }],
});
console.log(resp.choices[0].message.content);

Implement streaming in Node.js by iterating over the async generator. This approach handles partial responses efficiently, keeping your UI responsive even for longer outputs.

Use Tool Calling Functions

The proxy supports function calling for enhanced agent capabilities. Define your tools in the request payload and the model will return structured JSON for tool execution. This feature works seamlessly with the uncensored model, allowing for robust agent workflows without vendor lock-in.

Manage Your API Key and Limits

Your account includes one API key, which can be regenerated at any time. Note that regeneration revokes the old key immediately. Monitor your usage to avoid interruptions; requests return a 402 error when credit is insufficient and a 429 error if you exceed 300 requests per minute. The context window is 64,000 tokens for prompt plus completion.

Streaming

Copystream = client.chat.completions.create(
    model="uncensored",
    messages=[{"role": "user", "content": "Tell the story in second person."}],
    stream=True,
)
for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)