Integrate the uncensored chat completions API

Authentication and base URL setup

Start by generating your key via Google or email login on the Get API key page. The system issues one active key per account immediately, and you can replace it anytime. Use the base URL HTTPS://api.groqapikeys.com/v1 with your key as the bearer token. This configuration works with any OpenAI-compatible client without extra proxy steps.

Copycurl https://api.groqapikeys.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "List three advantages of low latency inference."}]
  }'

Chat completions endpoint structure

Send POST requests to /v1/chat/completions with your model id as "uncensored". The endpoint accepts standard message arrays and returns text output. We do not offer embeddings, image generation, or fine-tuning. Keep your payloads under 8 MB to ensure reliable processing. Errors like 401 (invalid key) or 402 (no credit) are returned instantly so you can adjust your integration.

Streaming responses and token counting

Enable streaming by setting the stream parameter to true in your requests. You receive a sequence of Server-Sent Events (SSE) chunks that allow you to display text progressively. The final chunk contains the complete usage statistics for the request, including input and output token counts. This approach reduces perceived latency for large responses and keeps your UI responsive.

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)

Tool calling and JSON mode configuration

Configure structured outputs by setting the response_format to json_object. This forces the model to return valid JSON instead of free text. You can also define tools in your request body to enable function calling capabilities. The uncensored model handles complex tool definitions reliably without refusing standard JSON structures. Use this for data extraction or API integration tasks.

Copyimport OpenAI from "openai";

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

Rate limits and concurrency constraints

Each API key allows up to 300 requests per minute and supports 8 concurrent requests. The context window spans 100,000 tokens for both prompt and completion combined. If you exceed these limits, the system returns a 429 rate limit error. Keep your request bodies under 8 MB to avoid processing delays. These limits ensure stable performance for high-throughput applications without hidden tier restrictions.

Error handling and usage reporting

Monitor your account balance via the dashboard since credit is prepaid and never expires. Requests that fail due to content refusal or network issues do not consume your tokens. If you encounter a double charge or billing issue, use the Support page to resolve it. The model does not refuse lawful adult or controversial topics, so most 4xx errors relate to authentication or limits.

Copyfrom openai import OpenAI

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