Animate a still frame with Kuaishou's Kling v2.1.
Model Category
Image-to-Video Models — 120+ Ways to Animate a Still Frame
The largest category in the catalog — 120+ image-to-video models turn a still frame into motion, from a simple animate pass to Seedance 2.0 I2V's 9-reference-image control.
120+ Image-to-Video Models9 Reference Images (Seedance 2.0 I2V)
Image-to-video animates a still frame — the one you just generated in Image Studio, or any upload — into motion. At 120+ models it's the single largest category in the whole catalog, covering everything from a simple animate-this-photo pass to multi-reference-image motion control.
Image-to-Video Models
120+ Image-to-Video Models, One Open Generative AI Catalog
Every image-to-video model below runs through the same Video Studio panel — upload a start frame and the model set switches automatically, with no separate accounts and no filters swapped in per model.
Google Veo 3's animation mode.
Runway's animation model.
ByteDance — up to 9 reference images, 4 aspect ratios, 5/10/15s duration.
Midjourney's animation mode.
Tencent's Hunyuan — also runs locally via Wan2GP.
Alibaba Wan 2.2 — hosted, or local through Wan2GP on a CUDA/ROCm GPU.
xAI — cinematic motion from a still image, up to 15s.
Hosted vs. run it locally
Animate in the Cloud, or on Your Own GPU
Open Generative AI ships the same image-to-video catalog two ways — a hosted app with zero setup, and a self-run Wan2GP server that animates your stills on your own machine.
Hosted
The hosted catalog covers 120+ image-to-video models, the largest category in the app, from Kling v2.1 I2V to Seedance 2.0 I2V's 9-reference-image animation.
Run It Locally
Wan 2.2's image-to-video mode and Hunyuan Video run locally through Wan2GP on a self-run CUDA or ROCm GPU server; the desktop app just sends prompts and receives results over HTTP.
Compare
Three Image-to-Video Models, Side by Side
A quick look at three representative models in the catalog — pick the one that matches the motion you need.
| Model | Provider / Source | What it's built for |
|---|---|---|
| Kling v2.1 I2V | Kuaishou | Animate a still frame into video |
| Seedance 2.0 I2V | ByteDance | Up to 9 reference images, 4 aspect ratios, 5/10/15s duration |
| Wan 2.2 I2V | Alibaba (Wan) | Runs hosted, or locally via Wan2GP on a CUDA/ROCm GPU |
Swipe to see the full table
FAQ
Image-to-Video Model FAQ
120+ — the largest single category in the catalog — including Kling v2.1 I2V, Veo3 I2V, Runway I2V, Seedance 2.0 I2V, Midjourney v7 I2V, Hunyuan I2V and Wan 2.2 I2V.
Explore more
More Open Generative AI Model Categories
This category pairs closely with the ones below, plus the Video Studio page that runs it end to end.
Animate a Still With Open Generative AI — Free, Right Now
120+ image-to-video models, no content filters, no subscription. No account required to try it.
Try Free