Wan 2.6

• 🎬 Alibaba's image-to-video model animating a still first frame.
• 👁️ Takes a first-frame image plus a text prompt as input.
• 🎯 Photorealistic generation with strong subject coherence across frames.
• 🔧 Available through Alibaba Cloud's Model Studio API.
• 🏢 Part of Alibaba's Wan (WanXiang) generative video suite.
• 🆕 Released December 2025; later succeeded by Wan 2.7.

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 2.6 is the image-to-video member of Alibaba's Wan family, taking a single first-frame image together with a text prompt and producing a video clip from it. It is offered through Alibaba Cloud's Model Studio platform, which also documents the companion Wan 2.6 image-generation endpoint. The catalog positions it as a photorealistic generator with strong subject coherence across frames.

Within the family, Wan 2.6 follows the earlier Wan 2.5 Preview image-to-video model and shares its release generation with the companion Wan 2.6 text-to-video model and the faster Wan 2.6 Flash variant.

The 2.6 generation focuses on first-frame-to-video, while the later Wan 2.7 forms the next generation in Alibaba's Wan video lineup. Wan 2.6 remains a usable option for animating existing image assets and storyboard frames through the Model Studio API.