Connect in 3 lines of code

Base URL and Model ID

Your first step is pointing your client at our server. Use https://api.openaiapialternatives.com/v1 as the base URL and set the model ID to uncensored. This single endpoint handles all chat completions without requiring extra configuration for embeddings or audio. You can verify connectivity by hitting the models endpoint to confirm the system is active and ready for your requests.

Authentication Header

Send your API key in the Authorization header using the Bearer scheme. Generate your key instantly via Google or email on the key page with no phone number required. The key grants access to your prepaid credit balance, so keep it secure. If you lose it, generate a new one to replace the old active key immediately.

Standard Chat Completions

Make a standard POST request to /v1/chat/completions with your message array. Our model responds with text directly, supporting parameters like temperature and top_p for control. Errors such as invalid keys return 401, while insufficient credit triggers a 402. Use JSON mode for structured outputs or stick to plain text for conversational flows.

Streaming Responses

Enable streaming by setting stream: true in your request payload. You receive Server-Sent Events (SSE) chunks as the model generates text. The final chunk contains the total token usage for billing accuracy. This approach keeps your UI responsive while waiting for long-form answers from the uncensored model.

Function Calling Tools

Define tools in your request to let the model output structured data. Pass the tools array with function definitions, and the model will return tool calls when appropriate. You can enforce specific tool usage with tool_choice or let the model decide. This feature works seamlessly with the standard OpenAI SDKs for agent workflows.

Limits, Errors, and Context

Your context window supports 100,000 tokens total, with a max output of 16,000 tokens per request. Rate limits cap you at 300 requests per minute and 8 concurrent requests per key. Handle 429 errors by implementing exponential backoff. Requests fail gracefully if credit runs out, and unused prepaid credit never expires.

cURL

Copycurl https://api.openaiapialternatives.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "Translate the following French code comments to English."}]
  }'

Python

Copyfrom openai import OpenAI

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