polym-eval-generate-gpt-image

What problem does it solve?

This Skill helps you quickly turn text prompts into images and perform prompt-guided edits on existing images, including targeted masked changes, without manually calling the underlying AIDP endpoints.

Core Features & Use Cases

• Text-to-image generation: Create new images from prompts using the gpt-image-1 model through ByteDance AIDP.
• Image editing/compositing: Apply a new prompt to one or more reference images using AIDP’s edits endpoint.
• Masked edits: Provide a mask image to limit where changes should occur for more precise results (e.g., background replacement or object retouching).
• Practical use case: Produce multiple variations of a concept image, then refine specific regions (like swapping the sky or dressing a subject) by running edit mode with an input image and an optional mask.

Quick Start

Run the skill in generate mode with your prompt to produce an image.