Integration guide for the uncensored endpoint

Base URL and authentication

Start by pointing your OpenAI-compatible client at our endpoint. Replace the standard base URL with https://api.llmgatewayopensource.com/v1 and swap your API key into the authorization header. This single change redirects traffic to our uncensored model without touching your application logic. You generate keys via Google or email on the dashboard. Each account holds one active key that replaces the previous one instantly. No configuration files or complex routing layers are required.

Making your first request

Send a standard chat completion payload to test connectivity immediately. The endpoint accepts text input and returns generated text without content refusals for lawful adult use. You can verify the response by checking the generated text field. Errors like invalid keys return 401, while insufficient credit returns 402. Use this simple command to confirm your setup works before integrating it into your main workflow.

Copycurl https://api.llmgatewayopensource.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "Fix the syntax error in this Python loop."}]
  }'

Using the Python SDK

Install the official library and initialize it with our base URL and your key. This approach lets you reuse familiar code patterns while accessing the uncensored model. The client handles serialization and streaming automatically. You can set parameters like temperature or top_p to control creativity. This method is ideal for backend services or scripts that need reliable text generation. Pass the model ID uncensored to ensure you hit the correct endpoint.

Copyfrom openai import OpenAI

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

Integrating with Node.js

Initialize the client using the base URL and your API key in your Node environment. This setup supports both standard JSON responses and streaming events. You can pass system prompts or user messages just like you would with other vendors. The library manages the connection lifecycle efficiently. This is the preferred method for web servers or serverless functions that need to generate text on demand. Ensure your code handles the response object correctly to extract the content.

Copyimport OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.llmgatewayopensource.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 with SSE

Enable streaming by setting the stream parameter to true in your request. The server sends chunks of text as they are generated, allowing for lower latency perception. Each chunk includes token usage data in the final piece. This feature is essential for user interfaces that display text progressively. You can parse the Server-Sent Events stream to update the UI in real time. It reduces the wait time for long outputs significantly.

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)

Limits, errors, and context

Our API enforces strict but fair limits to ensure stability. You get 300 requests per minute and a concurrency of 8 simultaneous requests per key. The context window supports 100,000 tokens for both prompt and completion combined. Max output per request is 16,000 tokens unless you specify a lower limit. Errors include 401 for bad keys, 402 for low credit, and 429 for rate limits. Request bodies must stay under 8 MB. These constraints are transparent and consistent.