Quickstart: Connect Your Client

Base URL & Authentication

Our API follows the standard OpenAI format, allowing you to switch providers by changing only two variables: the base URL and the API key. You can obtain a key by signing up on the Get API key page with just an email and password. No credit card is required for the trial, and keys can be regenerated instantly if compromised.

Use the base URL https://api.chinesellmapi.com/v1 in your client configuration. The model ID you must send is uncensored. This is an open-weight model running on our dedicated infrastructure, distinct from GPT, Claude, or other vendor models. It is tuned to answer without content refusals for lawful adult use.

Chat Completions Endpoint (POST /v1/chat/completions)

Send text in and receive text out using the standard chat completions endpoint. This endpoint supports both non-streaming and streaming responses, as well as tool/function calling. You must include the model field set to uncensored in your request body.

Requests are limited to 8 MB in body size. If you exceed the 100,000 token context window (prompt + completion), the API will reject the request. Ensure your client handles standard HTTP responses correctly.

• Input: JSON body with messages array.
• Output: JSON with choices containing message.content.

Make your first request with this example:

Copycurl https://api.chinesellmapi.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."}]
  }'

Python SDK Integration

Using the official OpenAI Python SDK is the most straightforward way to integrate. Install the package via pip and configure the client to point to our base URL. This approach works seamlessly with any library built on the OpenAI Python client.

Remember that our model is not GPT-4 or any other vendor model. It is an uncensored model. Adjust your prompt engineering accordingly, as the model may have different stylistic tendencies compared to other models.

Here is how to initialize the client and send a request:

Copyfrom openai import OpenAI

client = OpenAI(base_url="https://api.chinesellmapi.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)

Node.js SDK Integration

For JavaScript or TypeScript projects, the openai npm package works with minor configuration changes. Set the baseURL to our API endpoint and provide your API key. This ensures compatibility with existing codebases that expect OpenAI-standard responses.

The response structure remains identical to the OpenAI format, including id, object, created, and choices fields. This makes it easy to swap in our uncensored model for testing or production without rewriting your data parsing logic.

Configure and call the API:

Copyimport OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.chinesellmapi.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);

Streaming Responses (SSE)

Enable streaming by setting stream: true in your request. The API returns Server-Sent Events (SSE) containing incremental chunks of the response. This is ideal for user interfaces that need to display text in real-time as it is generated.

Each chunk contains partial content in delta.content. Handle these chunks incrementally in your client to update the UI. The stream ends when the final chunk is received with finish_reason set to stop or tool_calls.

Example streaming implementation:

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)

Rate Limits, Errors & Constraints

Monitor your usage to avoid interruptions. Each API key is limited to 300 requests per minute. If you exceed this, you will receive a 429 Too Many Requests error. Request bodies must not exceed 8 MB. The context window is strictly 100,000 tokens for prompt plus completion combined.

Common errors include:

• 401 Unauthorized: Invalid or missing API key.
• 402 Payment Required: Insufficient prepaid credit.
• 429 Rate Limit: Exceeded 300 requests per minute.

Pricing is transparent: $0.25 per 1M input tokens and $1.00 per 1M output tokens. Credits do not expire. Use the GET /v1/models endpoint to verify available models, though only uncensored is offered for chat completions.