Compare Wan2.2 vs. Wan2.5

Wan2.2 marks a significant enhancement to the Wan suite of open video foundation models by incorporating a Mixture-of-Experts (MoE) architecture that separates the diffusion denoising process into high-noise and low-noise pathways, allowing for a substantial increase in model capacity while maintaining low inference costs. This upgrade leverages carefully labeled aesthetic data that encompasses various elements such as lighting, composition, contrast, and color tone, facilitating highly precise and controllable cinematic-style video production. With training on over 65% more images and 83% more videos compared to its predecessor, Wan2.2 achieves exceptional performance in the realms of motion, semantic understanding, and aesthetic generalization. Furthermore, the release features a compact TI2V-5B model that employs a sophisticated VAE and boasts a remarkable 16×16×4 compression ratio, enabling both text-to-video and image-to-video synthesis at 720p/24 fps on consumer-grade GPUs like the RTX 4090. Additionally, prebuilt checkpoints for T2V-A14B, I2V-A14B, and TI2V-5B models are available, ensuring effortless integration into various projects and workflows. This advancement not only enhances the capabilities of video generation but also sets a new benchmark for the efficiency and quality of open video models in the industry.

Wan2.5-Preview arrives with a groundbreaking multimodal foundation that unifies understanding and generation across text, imagery, audio, and video. Its native multimodal design, trained jointly across diverse data sources, enables tighter modal alignment, smoother instruction execution, and highly coherent audio-visual output. Through reinforcement learning from human feedback, it continually adapts to aesthetic preferences, resulting in more natural visuals and fluid motion dynamics. Wan2.5 supports cinematic 1080p video generation with synchronized audio, including multi-speaker content, layered sound effects, and dynamic compositions. Creators can control outputs using text prompts, reference images, or audio cues, unlocking a new range of storytelling and production workflows. For still imagery, the model achieves photorealism, artistic versatility, and strong typography, plus professional-level chart and design rendering. Its editing tools allow users to perform conversational adjustments, merge concepts, recolor products, modify materials, and refine details at pixel precision. This preview marks a major leap toward fully integrated multimodal creativity powered by AI.