One API for All AI Models

TL;DR apimodels.app is a unified AI model API: one API key calls image, video, LLM and audio models (Seedance, Grok Imagine, Kling, gpt-image-2, nanobanana, Claude, Gemini and more) — just switch the model parameter. No separate provider accounts, 60-95% cheaper than official, pay-as-you-go (failed tasks not charged), reachable anywhere, $1 free on signup.

Unified endpoints

Type	Endpoint	Example models	Docs
Image	POST /api/v1/images/generations	gpt-image-2 · nanobanana · kling-v3-image	Docs
Video	POST /api/v1/video/generations	seedance-2.0 · grok-imagine-video-1.5 · kling	Docs
LLM	POST /api/v1/messages	Claude · Gemini · GPT	Docs
Audio	POST /api/v1/audio/generations	ElevenLabs · Kling Audio · Minimax	Docs

Quick start

curl -X POST https://apimodels.app/api/v1/images/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model": "gpt-image-2", "prompt": "a cute cat" }'

See all models and pricing: Pricing · Models · API Docs

Frequently Asked Questions

Can I call multiple AI models with one API?

Yes. apimodels.app provides a unified API — one API key calls image, video, LLM and audio models; just switch the model parameter, no separate provider accounts needed.

Which models are supported?

Seedance 2.0, Grok Imagine Video 1.5, Kling, gpt-image-2, nanobanana (Gemini-family image), Claude, Gemini, GPT, ElevenLabs and more, updated continuously. See the full pricing table for the current list.

Is it cheaper than the official APIs?

60-95% cheaper. Pay-as-you-go, failed tasks not charged, $1 free on signup, Stripe and Alipay supported.

Can I use it from anywhere, including China?

Yes — the unified API is directly reachable, so you do not need to handle each provider's regional network access.

How do I start?

Sign up at apimodels.app → get an API key → POST to the right endpoint (image/video/LLM/audio) with the model name. cURL / Python / Node.js examples are provided.