Browse 1,000+ AI Models
Discover and experiment with 1,000+ AI models spanning image generation, video, audio, 3D, avatars, creative tools, and more—all in a unified platform. Instantly try any model in your browser, or integrate them into your products via a simple, unified API. Developers can access detailed API docs, playgrounds, and code samples for seamless integration. Enjoy generous free credits on sign up, with no credit card required, so you can explore model capabilities risk-free before upgrading. Begin building with state-of-the-art AI in minutes!
Ultimate AI Media
Generation Platform
Skywork AI Models
Mureka AI Models
Clarity AI Models
Pruna AI Models
HappyHorse Models
Seedance 2.0 Models
Wan 2.7 Models
Qwen Image 2 Models
Grok Models
Seedance 1.5 Pro Models
Wan 2.6 Models
Kling O3 Models
OpenAI Models
Wan 2.5 Models
Seedream Models
Wan 2.2 Models
Dreamina AI Models
Seedance Models
Flux Image Tools
Minimax Hailuo Models
Kling Models
Google Models
Flux Kontext Models
Runwayml AI Models
Wan 2.1 Video Models
Hunyuan Models
Vidu Models
Ideogram Image Models
Recraft Image
Qwen AI Models
Pixverse AI Models
Stability AI Models
Video Extend
Object Detection and Segmentation
Content Detection Models
Motion Control Models
Best Video Models
Best Image Models
Swap Anything
Audio for Video
Video Edit
Ultra Selection
LoRA Generation
Generate Music
First and Last Frame Video
Remove Anything
3D Creation
Avatar Lipsync Models
Training Tools
Enhance Videos
Image Editing
Upscale Image
Speech Generation
Per-model API landing pages
WaveSpeedAI Models — Frequently asked questions
What is WaveSpeedAI?
WaveSpeedAI is an AI inference platform that exposes 1,000+ image, video, audio, and 3D models from providers like ByteDance, Google, Alibaba, OpenAI, Kuaishou, ElevenLabs, and more — all behind a single REST API and one billing account. No per-provider setup, no separate API keys, no per-vendor rate-limit juggling.
How do I use a model on WaveSpeedAI?
Pick a model from the catalog above, open its playground page, then either run it interactively or copy the auto-generated Python / JavaScript / cURL example into your own code. Submission returns a prediction id; poll the prediction endpoint until status is "completed" and read the output URL from data.outputs[0]. Full docs at wavespeed.ai/docs.
How is pricing calculated?
Each model lists a base price per run. The final charge scales with the parameters you set (resolution, duration, output count, references — whichever knobs that model exposes), and the exact cost for your current input is shown live next to the Generate button before you submit. Pricing is per output, not per second of compute, so it stays predictable as you scale.
Do I need an API key to try a model?
You can run any model from the in-browser playground without writing code. Once you want to integrate the API into your own app you create an API key from /accesskey — new accounts get free starter credits.
Which model categories are available?
Image generation (text-to-image, image editing), video generation (text-to-video, image-to-video, video editing, video extend), motion control + digital humans, 3D asset generation, audio + speech synthesis, upscalers, watermark / object removal, and 1,000+ LoRA variants. Filter the catalog by type using the pills above the grid.
Can I run multiple models with one account?
Yes — one API key, one billing account, one rate-limit envelope across every model. Cross-vendor workflows (e.g. Seedream → Seedance → upscaler) compose without juggling separate authentications.