Integrate the uncensored chat-completions API

Base URL and Authentication Setup

The API lives at https://api.apillmsourceopen.com/v1. Point your SDK there and supply your key. You get one active key per account; generating a new one invalidates the previous key immediately. No phone number is required for signup. We support Google login or a standard email and password flow. Your prompts never leave our servers for training purposes. This setup works with any client that accepts the OpenAI V1 format.

Sending Your First Completion Request

Post messages to the chat endpoint to get immediate text back. The model id is always "uncensored". It handles adult themes and controversial topics without blocking lawful content. You do not need to specify a model version. The system processes your input and returns a response. Use the standard format for messages. This approach ensures compatibility with existing tooling.

Configuring Streaming and Token Limits

Stream responses by setting stream: true. You receive chunks as they generate. The final chunk contains the full token usage statistics. Context window support reaches 100,000 tokens for prompt and completion combined. Max output per request is 16,000 tokens. If you omit max_tokens, the default is 2,048 tokens. Streaming reduces perceived latency for long answers. It also allows progressive rendering in your interface.

Enabling JSON Mode and Function Calling

Use response_format: {"type": "json_object"} for strict JSON output. Function calling works via the standard tools parameter. The model supports tool_choice for explicit instruction. You can set temperature, top_p, and seed for control. Frequency and presence penalties are available too. Stop sequences work as expected. This flexibility allows complex integrations. Structured data extraction becomes reliable.

Handling Errors and Refusals Gracefully

Standard HTTP codes indicate status. A 401 means your API key is invalid or expired. A 402 error signals insufficient prepaid credit. A 429 response indicates you hit the rate limit. Errors are free; you do not pay for failed requests. The only hard refusal is for sexual content involving minors. Everything else passes through. Check your balance regularly. Retry logic should handle 429s with exponential backoff.

Understanding Rate Limits and Quotas

You get 300 requests per minute per key. Only 8 requests run simultaneously per key. The request body maxes out at 8 MB. One active key per account ensures clear usage tracking. A new key replaces the old one instantly. Prepaid credit charges by real token usage. No monthly fees apply. Credit never expires. This structure suits bursty workloads. Predictable costs help manage budgets. Crypto top-ups add bonus credit automatically.

cURL

Copycurl https://api.apillmsourceopen.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "Generate a 256 byte binary blob"}]
  }'

Python

Copyfrom openai import OpenAI

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