Wan2.2
Alibaba's open-source video model (Wan series) — text-to-video and image-to-video
Alibaba's open-source video generation model (the Wan series) for text-to-video and image-to-video, under continuous iteration.
Create and edit video with AI — generation, editing, and effects.
Key features
- Open large-scale video generative models with MoE architecture
- Cinematic aesthetics with controllable lighting and composition
- Text-to-video and image-to-video at 720P 24fps on consumer GPUs
- Animate and speech-to-video variants; Diffusers and ComfyUI support
Good for
- Generating AI video from text or images for research and products
- Character animation and audio-driven video generation
Stack & tags
video-generationtext-to-videoimage-to-videoopen-weightsdiffusionmoecinematicai-video
Similar open-source projects
MoneyPrinterTurbo
★ 127,353One-click short videos from a topic — script, voiceover, and subtitles included
Deep-Live-Cam
★ 96,892Real-time face swap tool — driven by a single photo