Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

Why is image generation limited to OpenRouter when my OpenAI provider has GPT Image models?

Open
  • Is something wrong?
#828 0 comments 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
58/100
Issue type
Feature
Clarity
Mostly clear
Activity status
Active
Tech stack
python, typescript
Domain
api, backend, frontend

Research direction

Start in ui-web/src/features/settings/providers/Roles.tsx, especially roleProviders() and the media-selection save path, then trace the media configuration schema in raven/config/schema.py and the provider handling in raven/agent/tools/media_gen.py. Verify the existing manual apiBase/apiKey behavior and identify how provider reuse should appear in Settings. Done means an existing OpenAI-compatible provider can be selected for image generation, its endpoint and credentials are used, and the relevant configuration path is tested or documented.

Written by the indexing model from the issue text.

Description

Question

Why does the image-generation model picker in Raven Settings require OpenRouter when an existing OpenAI provider already exposes GPT Image models? Could the image role use that provider directly?

Current configuration (credentials and endpoint omitted)

  • Raven v0.2.3.
  • agents.defaults.provider is openai (the chat model is openai/gpt-6-sol).
  • providers.openai has an API key and a custom OpenAI-compatible apiBase. That endpoint also offers a gpt-image-2.5 model.
  • tools.media.image currently contains { "model": "gpt-image-2.5", "quality": "" }; it has no separate apiKey or apiBase.
  • An OpenRouter provider is also configured, but I would like to use the image capability of the existing OpenAI-compatible endpoint without routing or configuring it as OpenRouter.

What I see

The Settings image-role picker offers only OpenRouter. On current main, roleProviders() filters media roles to openrouter, and saving a media selection writes only the model and quality. The backend borrows a missing media API key only from providers.openrouter, while an unset media apiBase defaults to OpenRouter.

The image tool itself already appears to support OpenAI-compatible /images/generations and /images/edits endpoints when tools.media.image.apiBase and apiKey are supplied explicitly. That makes this look like a Settings/provider-selection limitation rather than an Images API limitation. I have not tested the manual per-tool override yet.

Expected behavior / request

Could the image-generation role let me select an already configured OpenAI or OpenAI-compatible provider and reuse its apiBase and apiKey for GPT Image models? If there is a technical reason to keep the picker OpenRouter-only, could the intended configuration for this case be documented in Settings and the setup guide?

Dominant language
Python
Stars
4.1k
Forks
94
Avg merge
10h 2m
Merged PRs (30d)
376

Getting set up

This project ships no dev container, Dockerfile or contributing guide, so setting up is up to you: start from its README, and see our first-contribution guide for the general steps.

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from EverMind-AI/Raven

All issues in EverMind-AI/Raven

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.