Uncensored AI API Documentation

• Set role to user or assistant.
• Pass content as strings.
• Receive JSON responses.

Errors return standard HTTP codes. Check the error object for details. Retry logic should handle transient failures gracefully.

Model Details

The model ID is uncensored. It is an open-weight model tuned for minimal refusal. It handles controversial topics well. Context window spans 100,000 tokens total. Max output is 16,000 tokens. Default max output is 2,048 tokens. This allows long conversations or document processing. The model is not GPT or Claude. It is a distinct uncensored model. It runs on our servers. You pay per token used. Input costs $0.15 per 1M tokens. Output costs $0.50 per 1M tokens. Crypto payments only. USDT TRC20 or USDC Base. No subscriptions. Credit never expires. Top up from $20 to $500. Bonus credit applies. +5% from $50. +10% from $100. +15% from $200. Direct crypto only.

Streaming and Token Usage

Enable streaming by setting stream: true. You receive Server-Sent Events (SSE). Each chunk contains partial text. The final chunk includes token usage data. This allows real-time display of responses. It reduces perceived latency. Users see tokens appearing instantly. You can stop generation early if needed. The API supports graceful interruption. Token counts help track costs accurately. Input tokens are counted separately from output tokens. Monitor usage to stay within limits. 300 requests per minute allowed. 8 concurrent requests per key. 8 MB request body limit. One key per account. Replace old keys easily. Streaming works with all SDKs. Python and Node clients handle SSE well.

Function Calling and JSON Mode

Use function calling for structured outputs. Define tools in your request. The model returns structured data. Use response_format with {"type": "json_object"} for strict JSON. This ensures valid JSON responses. Useful for parsers and automated workflows. You can mix tools and JSON mode. Temperature affects randomness. Set seed for reproducibility. Use stop sequences to halt generation. Presence and frequency penalties control repetition. These features enable complex applications. Build agents that call APIs. Parse results reliably. JSON mode prevents formatting errors. Function calling allows multi-step reasoning. Combine both for robust systems. Test with small payloads first. Verify structure before scaling.

Rate Limits and Constraints

Respect the 300 requests per minute limit. You can have 8 concurrent requests. Request bodies must be under 8 MB. One active key per account. A new key replaces the old one. Errors do not consume credit. Refusals are free. Context window is 100,000 tokens. Max output is 16,000 tokens. Default output is 2,048 tokens. $0.50 trial credit lasts 7 days. Crypto top-ups start at $20. Bonus credits add value. No expiration on credit. Support page handles refunds. Mistakes like double charges are fixed. Privacy is maintained. Prompts not used for training. Email only for signup. No phone number required. Google login available. Simple and direct. No complex setups.

cURL

Copycurl https://api.leaderboardllmuncensored.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "List the top 3 uncensored models for roleplay."}]
  }'

Python

Copyfrom openai import OpenAI

client = OpenAI(base_url="https://api.leaderboardllmuncensored.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

Copyimport OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.leaderboardllmuncensored.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

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)