Choose flagship
Use the top model when correctness, reasoning depth, and coding reliability matter more than cost.
Wan collection
Use Wan models for video generation, image-to-video, and reference-video style production workflows.
| Model | Best For | Input | Output | Context | Cached |
|---|---|---|---|---|---|
Wan 2.7 Video | General video generation model for text, image, and reference-driven video tasks. | Text / image / video | Video | Async | Task API |
Use the top model when correctness, reasoning depth, and coding reliability matter more than cost.
Use value models for assistants, automation, support, and workflows that still need strong quality.
Use lightweight models for routing, classification, extraction, and high-volume background tasks.
Prototype and ship features on Video models without managing multiple provider SDKs.
Route across the family by changing the model id — pricing, channels, and fallbacks stay on the platform.
Evaluate versions side by side, then move traffic between models without rewriting integration code.
Unified access
Keep product code stable while changing models, providers, pricing, and channel routes from the platform.
curl https://apiany.ai/v1/videos/generations \
-H "Authorization: Bearer $APIANY_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "wan-2.7",
"prompt": "A cinematic product reveal with soft studio lighting",
"duration": "5"
}'Send requests to one OpenAI-compatible APIAny endpoint with a single API key, and set the public model id in the request body — no provider-specific SDKs required.
Usage is paid from prepaid APIAny credits at the rates shown in the comparison table. Credits never expire and there is no subscription.
Yes. Change only the model id in your request — endpoint, auth, and response format stay the same across the family.
Use the comparison table and the How to choose guidance above: pick flagship for quality, balanced for everyday production, and budget for high-volume tasks.