Uncensored llm api documentation for developers

Base URL and authentication setup

Use the provided base URL for all requests. Authentication requires an API key sent in the header. Sign up via Google or email to get your key instantly. The key identifies your account for usage billing. Prepaid crypto credit funds your requests. Errors and refusals do not consume credit.

[[]]

[[]]

POST /v1/chat/completions example

Send messages to the chat endpoint. The model id is "uncensored". You can adjust parameters like temperature. The response contains the generated text. Use the official SDKs for easier integration. Change the base URL to point to our service. Headers must include your API key. The request body defines the conversation history.

[[]]

[[]]

Streaming responses via SSE

Enable streaming by setting the stream parameter. You receive chunks of text as they are generated. This reduces perceived latency for long responses. Parse the Server-Sent Events format correctly. Token usage is reported in the final chunk. This method works well for real-time displays. It provides a smoother user experience overall.

[[]]

[[]]

Function calling and JSON mode

Define tools for the model to call. The model returns structured arguments for these tools. You can also enforce strict JSON output. Set the response format to json_object. This ensures predictable parsing for your application. Combine tools with streaming for dynamic behavior. The model supports both structured and conversational outputs.

[[]]

[[]]

Parameters: temperature, top_p, and stop

Control randomness with temperature and top_p. Higher values make responses more creative. Lower values make them more deterministic. Use the stop parameter to halt generation. This helps control response length and format. Seed values allow for reproducible results. Presence and frequency penalties adjust word repetition. Adjust these settings to fit your use case.

[[]]

[[]]

Limits, errors, and context window

The context window supports 100,000 tokens total. Max output is 16,000 tokens per request. Rate limits allow 300 requests per minute. Concurrent requests are limited to eight per key. A 401 error indicates an invalid API key. A 402 error means you have no credit. A 429 error signals that you are rate limited. Request bodies must not exceed 8 MB.

[[]]

[[]]

cURL

Copycurl https://api.uncensoredlocalllms.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "Calculate the area of a circle with radius 5"}]
  }'

Python

Copyfrom openai import OpenAI

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