API Quickstart Guide

Authentication: Using Your API Key

Send your key in the Authorization header. Start by creating an account with Google or email. Your key appears immediately. No phone number blocks your access. Use Bearer YOUR_KEY for every request. The system rejects invalid keys with a 401 error. Check your balance before heavy usage.

Endpoint: /v1/chat/completions

Post JSON to the chat endpoint. This standard route accepts your prompt and model name. Send the model ID uncensored for raw output. The base URL is https://api.llmwithoutlogin.com/v1. You get text responses directly. Here is a basic curl example:

Copycurl https://api.llmwithoutlogin.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "Get the latest npm package version"}]
  }'

Keep your requests simple. The system processes them quickly. Avoid complex nested structures initially.

Parameters: Temperature, Top P, and Stop

Control creativity with these settings. Set temperature to 0.7 for balanced output. Use top_p to filter low-probability tokens. Define a stop sequence to end generation early. These parameters shape the response style. Python SDK handles them easily:

Copyfrom openai import OpenAI

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

Adjust values to fit your specific use case. Higher temperature increases randomness.

Streaming Responses via SSE

Receive tokens as they generate. Set stream: true in your request. The API returns Server-Sent Events. Each chunk contains partial text. This reduces perceived latency significantly. Use this for long outputs:

Copyimport OpenAI from "openai";

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

Buffer the chunks in your application. Concatenate them to form the final text. This method is efficient for real-time interfaces.

Function Calling and JSON Mode

Structure your output precisely. Use response_format with {"type": "json_object"}. This forces the model to return valid JSON. You can also define tools for function calling. The system adheres to your schema. Define the tools in your payload. This ensures reliable parsing in your code.

Handling Errors and Token Usage

Monitor your request status codes. A 401 error means your key is invalid. A 402 error indicates insufficient credit. A 429 error signals rate limiting. You have 300 requests per minute. The context window is 100,000 tokens. Output is capped at 16,000 tokens. Check the last SSE chunk for usage stats. Errors consume no credit. Your prepaid balance remains intact.

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)