Uncensored LLM API Documentation

Authentication and base URL

Your API key authenticates every request to the base URL https://api.aiapisubscription.com/v1. Generate a key instantly via Google or email on the dashboard. Pass it in the Authorization: Bearer <key> header. The system rejects invalid keys with a 401 error. Prepaid credit deducts only on successful completions. Errors consume zero tokens. Top-ups accept USDT (TRC20) or USDC (Base) from $20 to $500. Bonus credit applies automatically at higher tiers.

Chat completions endpoint

Send text messages to POST /v1/chat/completions. The model ID is uncensored. It handles roleplay and adult themes without standard refusals. Set max_tokens to control output length. Requests exceeding 8 MB return an error. The endpoint supports standard parameters like temperature and top_p. You can include a seed for reproducible outputs. System prompts set the initial context. User messages drive the conversation. Assistant responses follow naturally. This works with any OpenAI-compatible SDK.

Copycurl https://api.aiapisubscription.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "Summarize the technical specs of the new GPU."}]
  }'

Streaming responses and tokens

Enable streaming by setting stream: true. The API returns Server-Sent Events (SSE). Each chunk contains partial text. Token usage appears in the final chunk. This allows real-time display of generated content. You do not wait for the full response. Large outputs stream smoothly. Monitor the usage field to track consumption. Errors during streaming still return a final error chunk. Streaming works with standard HTTP clients. Use this for chat interfaces or live code generation. It reduces perceived latency significantly.

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)

Function calling and tools

Define tools in the tools array. The model can call functions based on context. Set tool_choice to auto or a specific function. The API returns tool calls in the response. Parse the tool_calls field in the JSON. Execute the function on your side. Send the result back in the next message. This enables dynamic behavior. The model understands complex schemas. Use this for agents or data extraction. It integrates with standard OpenAI tool formats. No extra configuration is required beyond the schema.

Copyfrom openai import OpenAI

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

JSON mode configuration

Force structured output by setting response_format: {"type": "json_object"}. The model returns valid JSON only. This is ideal for parsing data. Disable markdown formatting in the response. Use this for API endpoints or database inserts. The schema is not strictly enforced but highly reliable. Set temperature lower for consistency. JSON mode works with streaming. Errors are rare if the prompt is clear. Use this for deterministic data extraction. It simplifies client-side parsing significantly.

Copyimport OpenAI from "openai";

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

Concurrent requests are limited to 8 per key. You can send 300 requests per minute. The context window is 100,000 tokens total. This includes both prompt and completion. Exceeding the limit returns a 429 error. Requests larger than 8 MB are rejected. Credit runs out triggers a 402 error. Top up via crypto to resume. Errors do not consume credit. The key is unique per account. Replace it to generate a new one. Monitor usage via the final chunk.