Uncensored LLM API Documentation

Authentication & Keys

Start by generating a key on our dashboard. You sign up with Google or an email and password, then copy the string immediately. Send it in the Authorization header as a Bearer token. This single key controls your quota and streaming limits.

Copycurl https://api.bestuncensored.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "List the top 5 rust libraries for async web servers"}]
  }'

One active key per account. If you generate a new one, the old key stops working instantly. Your prepaid credit covers every request, and unused funds never expire.

Chat Completions Endpoint

Send messages to the /v1/chat/completions endpoint using the standard OpenAI format. Our uncensored model processes your system instructions and user prompts without refusing lawful adult or controversial topics. We return pure text responses in the choices array.

Copyfrom openai import OpenAI

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

Keep your payloads under 8 MB. The model accepts up to 100,000 tokens for the combined prompt and completion. If you do not set max_tokens, the output caps at 2,048 tokens. We do not offer embeddings, image generation, or fine-tuning endpoints.

Streaming & Token Usage

Enable streaming by setting stream to true. The server returns Server-Sent Events for every generated token. You receive real-time output while the model builds the response. Token usage data appears only in the final chunk of the stream.

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)

Handle partial responses gracefully. If the stream drops, you can retry the request. Our uncensored API supports function calling and JSON mode for structured outputs. Use response_format with json_object for consistent data types.

Parameters & Configuration

Control randomness with temperature and top_p. Set seed for reproducible results. Use stop sequences to halt generation early. Adjust presence_penalty and frequency_penalty to influence word repetition.

Copyimport OpenAI from "openai";

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

These parameters work exactly like the OpenAI SDK. We do not add hidden modifiers. Your requests are processed directly by our uncensored model. You can combine tools with tool_choice for automated workflows.

Error Codes & Limits

Monitor your responses for standard HTTP status codes. A 401 error means your API key is invalid or missing. A 402 error indicates you have insufficient prepaid credit. A 429 error signals that you have exceeded your rate limits.

We allow 300 requests per minute per key. You can also send 8 concurrent requests simultaneously. If you hit these limits, pause briefly before retrying. Errors and refusals do not consume your tokens. Only successful completions charge your balance.

Models List

Query the /v1/models endpoint to verify available options. Our uncensored API currently serves one large language model. The model id is uncensored. It is an open-weight model tuned for direct answers without content filters.

Copycurl https://api.bestuncensored.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "List the top 5 rust libraries for async web servers"}]
  }'

We do not route between multiple vendors. You are using our specific uncensored model. It supports the full chat completion format. Check the response object for the exact model version and pricing details per 1M tokens.