How do you call the Free AI API? Full reference

This page documents every endpoint of the free, keyless AI API. If you prefer not to read it, download the integration prompt and let your AI assistant write the code.

Base URL

https://api.apiaifree.com/v1

All requests use HTTPS. Request and response bodies are JSON in UTF-8, except for /ask, which returns plain text by default. Cross-origin requests are allowed from any website.

Authentication

There is none. The API has no keys, accounts or tokens. An Authorization header is accepted and ignored, so libraries that require a key work with any placeholder value such as free.

Chat completions

POST /v1/chat/completions is the main endpoint. It takes a conversation and returns the next assistant message.

Request parameters

Example request

curl https://api.apiaifree.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "free-ai",
    "max_tokens": 300,
    "temperature": 0.7,
    "messages": [
      {"role": "system", "content": "You answer in one short paragraph."},
      {"role": "user", "content": "Why is the sky blue?"}
    ]
  }'

Example response

{
  "id": "chatcmpl-8vlcqzR9pSE5FVxh",
  "object": "chat.completion",
  "model": "free-ai",
  "choices": [{
    "index": 0,
    "finish_reason": "stop",
    "message": {"role": "assistant", "content": "Sunlight is scattered by air molecules…"}
  }],
  "usage": {"prompt_tokens": 74, "completion_tokens": 58, "total_tokens": 132}
}

The answer text is in choices[0].message.content. A finish_reason of length means the answer was cut by max_tokens.

Streaming

With "stream": true the response has the content type text/event-stream. Each event is a line starting with data: that carries a JSON chunk; the new text is in choices[0].delta.content. The stream ends with data: [DONE].

const res = await fetch("https://api.apiaifree.com/v1/chat/completions", {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: JSON.stringify({ model: "free-ai", stream: true,
    messages: [{ role: "user", content: "Tell me a short story." }] }),
});
const reader = res.body.getReader(), dec = new TextDecoder();
let buf = "";
while (true) {
  const { done, value } = await reader.read();
  if (done) break;
  buf += dec.decode(value, { stream: true });
  const lines = buf.split("\n"); buf = lines.pop();
  for (const line of lines) {
    if (!line.startsWith("data:") || line.includes("[DONE]")) continue;
    const text = JSON.parse(line.slice(5)).choices?.[0]?.delta?.content;
    if (text) process.stdout.write(text);
  }
}

Reasoning mode

By default the model answers straight away. Send "reasoning": true to let it work through the problem first. Thinking consumes output tokens, so set max_tokens to at least 2000; with a small limit the visible answer can come back empty. The thinking text is returned separately in choices[0].message.reasoning_content.

The /ask endpoint

GET /ask or POST /ask answers a single question and returns plain text. It exists for shell scripts, spreadsheets, webhooks and other places where building JSON is inconvenient.

curl "https://api.apiaifree.com/ask?q=Give+me+a+synonym+for+fast"
curl "https://api.apiaifree.com/ask?q=Name+three+colors&format=json"

Text completions

POST /v1/completions continues a raw string passed in the prompt field, without chat roles. It accepts the same sampling parameters and returns the text in choices[0].text. Use it for autocomplete-style tasks; for everything else prefer chat completions.

Models and health

GET /v1/models returns a list with one entry, free-ai. GET /health returns {"status":"ok"} with status 200 when the model server is reachable, and status 503 when it is not. Poll it before long batch jobs.

Limits

Errors and retries

Errors use the familiar shape {"error": {"message": "…", "type": "…", "code": 400}} together with a matching HTTP status.

A simple exponential backoff covers both 429 and 503. Set your HTTP client timeout to at least 120 seconds, because long answers take time to generate.

SDKs and tools

Any client that speaks the chat completions format can use the API by overriding the base URL.

Python SDK

from openai import OpenAI
client = OpenAI(base_url="https://api.apiaifree.com/v1", api_key="free")

JavaScript SDK

import OpenAI from "openai";
const client = new OpenAI({ baseURL: "https://api.apiaifree.com/v1", apiKey: "free" });

LangChain, n8n and other tools

Choose the “OpenAI-compatible” or “custom base URL” option, enter https://api.apiaifree.com/v1, type any text into the key field and set the model to free-ai.