Skip to main content
Wan 2.7 is an image-to-video generation model. It takes a reference image and text prompt as the core creative inputs, then produces a video clip that follows the subject, framing, and visual direction established by the source image. The model is suited for short-form product motion, character or object animation, visual concept exploration, and social video assets where an existing image should become the anchor for camera movement and scene dynamics. Wan 2.7 supports 720p and 1080p output, 2-15 second durations, optional audio input, negative prompts, prompt rewriting, and reproducible seeds. In GENGEN, it uses the same canonical video task endpoint as Seedance models and returns wan27:<remote_id> public task IDs.

​
Model ID

Pass this exact public ID in the top-level model field. Do not substitute an upstream provider alias.

​
Supported modes

  • image_to_video

​
Inputs

  • Image
  • Audio
  • Text

​
Outputs

  • Video

​
Capabilities

  • Image-to-video

​
Additional capabilities

  • Image-to-video
  • Audio input
  • Prompt rewriting

​
Technical details

Availability and final customer prices can change. Check the Dashboard model catalog before production use.
See the API reference overview for authentication, request conventions, and response schemas.