fix: use gemini-2.5-flash-latest - #70
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: 6a2623ae24
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 2.5 Flash – tier 1) | ||
| # --------------------------------------------------------------------------- | ||
| MODEL_NAME = "gemini-2.5-flash-001" | ||
| MODEL_NAME = "gemini-2.5-flash-latest" |
There was a problem hiding this comment.
With a newly created or otherwise unprovisioned Gemini Developer API key, Gemini 2.5 Flash remains unavailable, so changing only the alias does not fix the observed not-found failures. Every non-dry call in deploy-docs.yml still passes this constant to client.models.generate_content, preventing the production AI update steps from completing; use a generally available model for this key, such as gemini-2.0-flash, or explicitly provision access before selecting 2.5 Flash.
Useful? React with 👍 / 👎.
1 participant
Previous attempt with
gemini-2.5-flash-001also failed (not found in v1beta API). Tryinggemini-2.5-flash-latestwhich should resolve to the current stable version.