fix: use gemini-2.0-flash (GA) - #71
Conversation
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 46ac075769
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| # Model & rate-limit constants (Gemini 2.0 Flash – tier 1) | ||
| # --------------------------------------------------------------------------- | ||
| MODEL_NAME = "gemini-2.5-flash-latest" | ||
| MODEL_NAME = "gemini-2.0-flash" |
There was a problem hiding this comment.
As of this commit's 2026-07-28 date, gemini-2.0-flash had already reached its Gemini API shutdown date on 2026-06-01. Every production generation and cache request passes this constant (for example, generate_content at lines 314/339 and caches.create at line 407), so the deploy-docs.yml update and image-description jobs will receive a model-not-found error rather than fixing access for the new API key. Select a currently served model instead.
Useful? React with 👍 / 👎.
1 participant
Why not 2.5 Flash
Google has restricted
gemini-2.5-flashto existing users only. The CI's API key is too new to access it. The-001suffix is a Vertex AI naming convention, not valid in the GenAI v1beta API.gemini-2.0-flashis GA (General Availability) and open to all API keys — this is the correct fix.