Skip to content

Fix automatic image generation on OpenAI-compatible gateways - #627

Open
drcxd wants to merge 1 commit into
editor-code-assistant:masterfrom
drcxd:fix-image-generation
Open

drcxd wants to merge 1 commit into
editor-code-assistant:masterfrom
drcxd:fix-image-generation

Conversation

@drcxd

@drcxd drcxd commented Sep 29, 2026

Copy link
Copy Markdown
  • I added a entry in changelog under unreleased section.
  • [?] This is not an AI slop.

Hello, this commit is intended to fix #612 I noticed that you dislike "ai slop", I do not like it either, but since I don't have the capability to fix this manually myself, and nobody is responding to that issue for 2 weeks. I tried asking an AI agent to fix it for me. Though I reviewed some of the changed code, it seems that image-generation is not enabled simply because the model is capable of that. Instead, it is only enabled for the exact OpenAI API URL, or if the configuration file explicitly enables that. I did not review the test though, but the agent reports that all tests are passed. If this is considered AI slop, please close the PR and I will no longer submit any PR like this.

Anyway, I build and test the project, and now I can use gpt-6-astral from our customized providers without the problem mentioned in the issue.

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

imagegen deployment must be provided through header: x-ms-oai-image-generation-deployment

1 participant