Summary
After a single prompt, Gemini CLI automatically switches from gemini-2.5-pro to gemini-2.5-flash, showing the quota-limit message even when only 1 query was executed.
Additionally, there’s a major delay in the Google Cloud Console API dashboard
where usage metrics have not been updating properly for several days — today the delay coincided with this quota-exhaustion problem.
Steps to reproduce
-
Open Gemini CLI (v0.9.0)
-
Run a single command such as: gemini "hello
-
Immediately receive: Automatically switching from gemini-2.5-pro to gemini-2.5-flash…
Possible reasons… daily gemini-2.5-pro quota limit -
Check Google Cloud Console → API Dashboard
→ notice extreme delay or missing updates for recent requests.
Expected behavior
Gemini 2.5 Pro should remain active for multiple queries within the daily Pro plan limits, and API Dashboard metrics should update normally.
Actual behavior
- CLI downgrades to Flash after the first prompt.
- Cloud Console metrics appear stale or delayed by several hours.
Environment
- OS: Windows 10 + WSL2 (Ubuntu 24.04)
- Gemini CLI version: 0.9.0
- Authentication: Gemini Advanced (Pro plan, personal Google account)
- Project: limiar-mcp-noticias
- Time observed: 22 Oct 2025 ~10:00 BRT
Additional context
Other users are reporting the same issue (e.g. GitHub #10518 and Forum #106962).
Please investigate if the Pro-quota check or Cloud Console metrics are temporarily desynchronized.