Wan2.2

Video AI Agents

Open-source video generation model family (5B/14B) for text/image/video-to-video, delivering 1080p clips with improved motion and control.

Wan2.2 is an open-weights family of large-scale video generative models released by the Wan-Video team as the successor to Wan2.1. It introduces expanded training data and architecture upgrades, offering multiple variants for text-to-video, image-to-video, video-to-video, and even speech-to-video generation. The models produce cinematic, higher-resolution clips with stronger temporal consistency and motion control, and are available under an Apache-2.0 license with runnable repos and Hugging Face checkpoints.

Industry: Marketing

Pricing: free

Use cases: video creators, VFX artists, game developers, researchers, creative technologists

Capabilities: Generating short cinematic videos directly from text prompts., Animating a single image into motion with image-to-video workflows., Converting/reference-editing existing video with video-to-video., Building ComfyUI or custom pipelines for creative production., Researching large video diffusion/MoE techniques with open weights.

Tags: text-to-video, image-to-video, video-to-video, speech-to-video, mixture-of-experts, temporal consistency, 1080p, ComfyUI, HuggingFace, Apache-2.0

• Does Wan2.2 support speech-to-video generation?
• Can Wan2.2 be integrated with ComfyUI workflows?
• Is Wan2.2 available under an Apache-2.0 license for commercial use?
• Are there any costs associated with using Wan2.2?