wan2.7-image-skill

AI image generation from text prompts requires inputting descriptive text into the Wan2.7-image model to automate creation. The skill handles resolution parsing and asynchronous API polling to deliver the final image URL for your creative projects.

Editing existing images using AI reference images is supported by uploading inputs via file_to_oss to the Wan2.7-image model. You can apply styles, merge multiple references, or transfer artistic directions onto new inputs through asynchronous task execution.

Creating a multi-image series for storyboards or tutorials uses the Wan2.7-image model to generate related visuals sequentially. You provide text prompts and optional references, and the asynchronous pipeline processes and delivers each final image URL.

Asynchronous task polling for AI image generation is handled automatically by the skill via API requests. You provide the prompt and optional OSS-uploaded reference images, and the system polls the Wan2.7-image model until task completion to return the final URL.

Managing resolution and input images for AI art involves using parse_resolution for dimensions and file_to_oss for uploads. The Wan2.7-image pipeline requires these inputs to properly configure the model before executing the asynchronous generation tasks.

AI image editing requires uploading reference images to OSS to provide accessible URLs for the Wan2.7-image model API. This enables the asynchronous processing pipeline to fetch and apply your reference styles or compositions during the generation task.