Self-hosted alternatives to openai/imagegen

O

openai/imagegen

Generate and edit images using OpenAI's Image API for projects

imagegen is an agent skill from OpenAI's curated skills collection for generating and editing raster images inside a Codex session. It is for tasks that benefit from AI-created bitmap visuals, such as photos, illustrations, textures, sprites, mockups and product shots, as well as transparent-background cutouts. It uses a built-in image tool by default, and it has a command-line fallback that calls the OpenAI image API, which runs only when you explicitly ask for it. The main advantage is staying in the same session as the code. Instead of switching to a separate image tool and describing the project again, you can generate and iterate on assets where the code lives. Typical uses include generating a hero image for a landing page with room for copy, replacing a product photo's background while keeping the product unchanged, producing several icon variants or sprite sheets for a game in one session, creating a UI mockup at a specified fidelity, and extracting a clean transparent cutout from a product photo. It suits developers and designers who need placeholder or final bitmap art for a project. Check the rights to any image you edit, and review generated images for accuracy before you publish them, since image models can make mistakes in details. The skill is listed on officialskills.sh and installs from the openai/skills repository. Install it with npx skills add https://github.com/openai/skills --skill imagegen.