Wan 2.7 Enhanced

• 🏢 Alibaba's Wan family video model, text-to-video generation
• 🎯 Photorealistic output with strong subject coherence across frames
• 🆕 Enhanced variant of the Wan 2.7 generation, released June 2026
• 🔧 Resolution-tier and aspect-ratio controls replace the older size field
• 💬 Shot type now driven by the prompt rather than a parameter
• 👁️ Image-to-video tasks include first-frame and video-continuation modes
• 🌐 Sits alongside base Wan 2.7 and image-to-video counterpart

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 2.7 Enhanced is a text-to-video model in Alibaba's Wan family, positioned as an enhanced build of the 2.7 generation that emphasizes photorealistic generation with strong subject coherence across frames. It sits alongside its own image-to-video counterpart, Wan 2.7 Enhanced, and follows the base Wan 2.7 in the series.

According to Alibaba Cloud's documentation, Wan 2.7 revises the API to use a resolution-tier and aspect-ratio combination in place of the older size field and drops the separate shot-type parameter in favor of prompt-driven shot description.

Compared with same-family predecessors Wan 2.6 and Wan 2.5 Preview, the 2.7 line adds multi-task video generation; Alibaba notes its new image-to-video API supports first-frame, first-and-last-frame, and video-continuation tasks, whereas the 2.6-and-earlier API handled first-frame only. These primary API references define what the 2.7 generation, and this Enhanced text-to-video variant within it, can be conditioned on.