Wan 2.6

• 🏢 Alibaba's Wan video family, image-to-video variant, released December 2025.
• 🎯 Animates a first-frame image plus a text prompt into video.
• 📏 Documented output at 720p or 1080p resolution.
• 🆕 Successor Wan 2.7 adds first-and-last-frame and video continuation.
• 🔧 Sibling text-to-video and a speed-tuned Flash variant exist.
• 👁️ Emphasis on subject coherence across frames and photorealism.
• 🌐 Available via Alibaba Cloud Model Studio and the DashScope SDK.

Alibaba

Alibaba Group is a Chinese multinational technology company founded in 1999 and headquartered in Hangzhou, Zhejiang. Originally built around e-commerce and cloud computing, Alibaba has become one of the most prolific contributors to open-weight AI research,…

Wan is Alibaba's flagship family of video generation models, and Wan 2.6 image-to-video is the photo-animation member of that lineup, released in December 2025. It takes a single first-frame image together with a text prompt and synthesizes a smooth clip, making it suited to product showcases, photo animation, and short visual storytelling. Per Alibaba Cloud's documentation, the Wan 2.6 image-to-video model supports first-frame-to-video generation, and API references list 720p or 1080p output.

Within the family, this variant pairs with the Wan 2.6 text-to-video model and a latency-optimized Wan 2.6 Flash option for faster iteration. It follows the earlier Wan 2.5 Preview in the same image-to-video lineage.

The generation that came after, Wan 2.7, is the more capable successor: Alibaba's documentation notes that Wan 2.7 image-to-video extends beyond first-frame animation to also support first-and-last-frame interpolation and video continuation, whereas Wan 2.6 and earlier handle first-frame-to-video only. Alibaba recommends Wan 2.7 as the preferred choice for new image-to-video work.

Across the broader Wan series, Alibaba's compare-and-select documentation lists multiple coexisting model tiers for different video generation tasks, with Wan 2.6 positioned for first-frame photo animation.