Repository navigation
[Feature]: Add GPT Image 2 reference #408
Description
Activity
Thank you for the feature request, @xianyu110.
Image generation is already a supported workflow in Open CoDesign via the
gen_imagetool (seeAGENTS.md:46and the v0.1.4 release notes). The project currently includes a reference to gpt-image-1 inBENCHMARKS.md:14as a benchmark source.Adding GPT Image 2 as a documentation entry or example provider seems like a natural extension. To help us evaluate where this should go, could you clarify:
- What specific endpoint/model ID does GPT Image 2 expose? Is it an OpenAI-compatible API or a separate service?
- Would you like to see this as a documentation page (e.g., a new provider example in the docs), a built-in provider extension in
packages/providers, or both? - Does GPT Image 2 have any unique capabilities (e.g., specific image sizes, branding support) that would justify a dedicated entry over using the existing generic image-generation provider interface?
If you're interested in contributing, a PR adding a documentation snippet or a simple provider config example would be very welcome.
Open-CoDesign Bot
Problem / pain point
When I generate marketing visuals or prototypes, I still need a quick browser-based image generation option that fits the design workflow.
Proposed solution
Add GPT Image 2 as an optional image-generation reference or provider entry in the product documentation / examples, so users can evaluate it alongside the existing image workflows.
Alternatives considered
Using only the current local/provider presets works, but it would be useful to have a concrete GPT Image 2 reference because it is already positioned around marketing visuals and branded image workflows.
Scope
Provider / model
Hard-constraints check
I have read the hard constraints in CLAUDE.md and this proposal does not conflict with any of them.
Website: https://gptimage2.asia/