KiosAPI – OpenAI-Compatible Gateway for GPT, Claude & Gemini

Access GPT, Claude, Gemini, and dozens of other AI models through one OpenAI-compatible endpoint and API key — no extra configuration needed.

KiosAPI is a unified API gateway that lets you call every major AI model — GPT-4o, Claude, Gemini, o1/o3, and more — through one endpoint using the OpenAI SDK you already know. Point your existing code at https://kiosapi.com/v1/ and swap your key. No code changes required.

Quickstart

Send your first request in under 5 minutes using the OpenAI SDK.

Authentication

Get your API key and learn how to pass it for each endpoint format.

API Reference

Full reference for chat, embeddings, moderation, and reranking endpoints.

Tutorials

Integrate KiosAPI with Claude Code, Codex CLI, Gemini CLI, and more.

How it works

KiosAPI supports three endpoint formats so you can use whichever SDK you already have installed. OpenAI-compatible libraries work via /v1/chat/completions, the Anthropic SDK works via /v1/messages, and the Google GenAI SDK works via /v1beta/models/{model}:generateContent — all under the same base URL and the same API key.

OpenAI Format

Use /v1/chat/completions with any OpenAI-compatible library.

Claude Native

Use /v1/messages with the Anthropic SDK directly.

Gemini Native

Use /v1beta/models/{model}:generateContent with the Google GenAI SDK.

Get started in 3 steps

1. Create your account

   Sign up at kiosapi.com and navigate to Dashboard → Token Management to generate your API key. It starts with sk-.

2. Point your SDK at KiosAPI

   Set base_url to https://kiosapi.com/v1/ and use your KiosAPI key wherever your SDK expects an API key. No other changes needed.

   quickstart.py

   from openai import OpenAI
   
   client = OpenAI(
       api_key="sk-xxx",
       base_url="https://kiosapi.com/v1/",
   )
   
3. Pick a model and send a request

   Pass any supported model name — gpt-4o, claude-sonnet-4-20250514, gemini-2.5-flash, and many more.

   quickstart.py

   response = client.chat.completions.create(
       model="gpt-4o-mini",
       messages=[{"role": "user", "content": "Hello!"}],
   )
   print(response.choices[0].message.content)
   

Explore the docs

OpenAI SDK Guide

Configure the OpenAI SDK for Python and Node.js with streaming examples.

High-Concurrency Requests

Run thousands of parallel requests using asyncio and aiohttp.

Token Quota

Check your spending limit and monitor usage via the billing API.

Function Calling

Use tool calling with any supported model via the OpenAI tools format.