GitHub radar
How to convert image to video in browser with Wan 2.2 14B? Video:
r3gm Hugging Face space, which allows you to generate 720P videos (at 24 FPS) from images and texts with Wan 2.2 without installing anything or signing up for anything.
Wan 2.2 is an open-sourced video generation model developed by the Wan-AI team. Given an input image and a textual description of the target motion, Wan 2.2 can generate 720P videos. In our r3gm Hugging Face space, we run Wan 2.2-14B under an optimized inference pipeline (fp8 quantization and AOT compilation) on free ZeroGPU resources with zero sign-ups required. Compared to Wan 2.1, Wan 2.2 is a huge improvement because it was trained on 83% more video data and has a new mixture-of-experts architecture that decomposes the generation task into different expert blocks to achieve smooth and coherent motion.
Why a vibe-coder should care
Playing around with such high-quality video models generally needs a high-end GPU and half an hour of setting up. Now you just need to browse, upload an image, tell us what to do, and wait for the magic.
How to install
Copy this and send it to your agent — Claude Code, Codex, any of them:
Open this space https://huggingface.co/spaces/r3gm/wan2-2-fp8da-aoti-preview and explain how to generate a video from a photo with a text prompt
Nothing to install — runs in the browser.
Open on Hugging Face▌ More finds