Pick Wan 2.6 or Wan 2.5
Use Wan 2.6 unless a prompt was already written for Wan 2.5.
AI Video
Wan AI (Wanxiang) is an Alibaba AI video generator for image to video and text to video. The versions here are Wan 2.6 and Wan 2.5. Upload a still and describe the motion, or write a shot from scratch, to get a short clip.
Use Wan 2.6 unless a prompt was already written for Wan 2.5.
For image to video, attach the frame and describe only the motion. For text to video, write subject, camera and light.
Length and aspect ratio are locked before the run.
Wan 2.6 is the current image-to-video entry. Upload a product photo, a character still or a key frame and describe only what moves. This is the usual path when the look is already decided.
Wan 2.5 is the previous named line. Keep it for prompts that were already tuned against that name.
Ask for a single camera move. A longer duration stretches the same action. For more events, generate more clips.
Wan AI is the Wanxiang video model for turning pictures or text into a short clip. Wan 2.6 and Wan 2.5 are the named versions here.
Yes. Upload a still and describe the motion. That is the usual way to keep a product or character looking the same in the clip.
Wan 2.6 is the current line. Wan 2.5 is the previous name. Keep 2.5 if the prompt was already tuned there.
Qwen Image makes stills. Wan AI makes video. Generate a still on Qwen Image if you need a first frame, then bring it to Wan.