Inference API: Quickstart Guide

Base URL and Authentication

Our API is fully OpenAI compatible. To switch, update your client configuration to point to our base URL and provide your API key. The base URL is https://api.llminferenceapi.com/v1. Authentication is handled via the standard Authorization: Bearer header. You receive your key immediately after signing up with just an email and password—no phone number or credit card required for the trial.

• Base URL: https://api.llminferenceapi.com/v1
• Header: Authorization: Bearer YOUR_API_KEY
• Model ID: uncensored

This setup works with any official OpenAI SDK or compatible client library. You keep your existing code structure; only the endpoint and key change.

First Request

Make your first call to the chat completions endpoint. The model ID is uncensored. This model is an open-weight large language model tuned to answer without content refusals for lawful adult use, not GPT or Claude.

Copycurl https://api.llminferenceapi.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "Write a blunt product review of a cheap VPN."}]
  }'

The request body includes the standard messages array. You can specify temperature, max tokens, and other standard parameters. The API returns a standard chat completion response with token usage statistics. If you encounter a 401 error, your key is invalid. A 402 error means your prepaid credits are exhausted.

Python SDK Integration

Using the Python SDK is straightforward. Initialize the client with our base URL and your API key. The code structure mirrors the standard OpenAI pattern. This allows you to swap in our llm proxy service without rewriting your application logic.

Copyfrom openai import OpenAI

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

Ensure you pass the model='uncensored' argument. The response object contains the generated text and usage metrics. This approach is ideal for developers who want a drop-in openai compatible api solution. No additional dependencies are required beyond the standard SDK.

Node.js SDK Usage

For Node.js developers, the integration is equally simple. Set the baseURL and apiKey in your client configuration. The SDK handles the HTTP requests to our ai chat api endpoints automatically.

Copyimport OpenAI from "openai";

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

Call chat.completions.create() with the model ID uncensored. You can stream responses or handle them as complete objects. This flexibility supports both batch processing and real-time user interactions. The API supports tool calling, allowing you to define functions that the model can invoke.

Streaming Responses

For real-time user experiences, enable streaming. Set the stream parameter to true in your request. The API returns a Server-Sent Events (SSE) stream of partial responses.

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)

Each chunk contains a partial delta of the response. You can render these tokens to the user as they arrive, reducing perceived latency. This is critical for chat interfaces where immediate feedback is expected. The streaming response includes the same token usage information as non-streaming requests.

Limits, Errors, and Context

Our API enforces a 64,000 token context window for both input and output combined. This supports long documents and complex conversations. Rate limiting is set to 300 requests per minute per key. If you exceed this, you receive a 429 error. Request bodies are limited to 8 MB.

Common errors include 401 (invalid key), 402 (insufficient credits), and 429 (rate limit). Credits never expire, so you can top up at your convenience. Prepaid credits start at $10, with bonuses for larger deposits. This pay as you go ai api model ensures you only pay for what you use.