Every state-of-the-art AI model behind one API.
One OpenAI-compatible endpoint for video, image, music and text. 300+ models from 19 providers, one API key, one bill — no subscription, pay per call.
One API. Four superpowers.
The same Bearer token works for generative video, images, music and chat. Change one parameter, get a different modality.
Video
Veo 3.1 · Runway Aleph · Seedance 2. Text-to-video, image-to-video, edit existing clips. 720p–1080p, up to 10 seconds.
Image
Flux Kontext Pro · Nano Banana · 4o Image. Text-to-image, editing, upscaling. PNG / JPG, any aspect ratio.
Music
Suno v3.5 → v4.5 Plus · MusicGen · Mureka. Up to 8-minute tracks with vocals, custom lyrics, persona cloning.
Chat
GPT-5 · Claude Opus 4.6 · Gemini 3 Pro. Long-context reasoning, tool use, streaming responses out of the box.
Hand-picked from the catalog.
Click any to see pricing, sample code and try it live in the playground.
Google nano banana 2, 1K
Google nano banana 2, 2K
Suno, Mashup
Suno, add-vocals
Runway Aleph
Google veo 3.1, Extend, Fast
Ship in 30 seconds.
Same request shape as the OpenAI-compatible providers — just change the host. Rate limit, errors, headers all identical.
curl https://aimarcusimage.eu/api/v1/jobs/createTask \
-H "Authorization: Bearer sk-aig-..." \
-H "Content-Type: application/json" \
-d '{
"model": "google/nano-banana",
"input": {
"prompt": "A golden retriever running through autumn leaves, cinematic",
"output_format": "png",
"image_size": "16:9"
}
}'
# → { "code": 200, "data": { "taskId": "..." } }
# Poll /api/v1/jobs/recordInfo?taskId=... until state="success"
The model labs you already trust.
Everything you need to know.
What is AI Generate? +
AI Generate is a single-API gateway to 300+ AI models across four modalities — video (Veo 3.1, Runway Aleph, Seedance 2), image (Flux Kontext Pro, Nano Banana, 4o Image), music (Suno), and chat (GPT, Claude, Gemini). One OpenAI-compatible key, one bill across every provider, and you pay only for what you call — no monthly fee.
How is the pricing calculated? +
You buy credits once ($10 / $50 / $200 packages) and spend them as you call the API. Each model has a per-unit price (per second of video, per image, per million tokens). See pricing for the full table, which is synced daily from upstream providers.
Is it OpenAI-compatible? +
Yes. The /api/v1/chat/completions endpoint accepts the standard OpenAI request shape (model + messages). Streaming SSE works out of the box. Just change the host and use your sk-aig-... key.
Can I use Claude or Gemini? +
Yes. We route Claude (claude-opus-4.6, claude-sonnet-4.5, claude-haiku-4.5) directly to Anthropic and Gemini through the OpenRouter / Google path. All with the same Bearer-token call.
How fast are image and video generations? +
Image generation (Nano Banana, Flux) typically completes in a few seconds; video (Veo Fast, Seedance Fast) in roughly 30–120 seconds depending on duration and resolution. Results arrive via webhook or polling.
Do you offer free credits? +
Every verified account gets a trial credit to make its first call for free. After that you top up from $10 — credits never expire and there is no subscription.
Is there a rate limit? +
20 generation requests per 10 seconds per API key. Polling task status does not count against the limit. Higher limits available for enterprise accounts.
Can I set spending limits? +
Yes. Daily and monthly USD caps configurable per account. API calls return HTTP 402 when the cap is reached until the next period, so you never get surprise bills.
Your first generation is free.
Sign up, verify your email, and make your first generation free — then pay only for what you call.