Skip to content

fix "minimal" thinking mode to "low" for gemini flash models - #281

Open
nmatton wants to merge 2 commits into
theJayTea:mainfrom
nmatton:fix_gemini_calls
Open

fix "minimal" thinking mode to "low" for gemini flash models#281
nmatton wants to merge 2 commits into
theJayTea:mainfrom
nmatton:fix_gemini_calls

Conversation

@nmatton

@nmatton nmatton commented Aug 31, 2026

Copy link
Copy Markdown

The gemini flah model series now have different thinking modes (thinking_level= “low”|”medium”|”high”) while current implementation uses "minimal", which is not included.

This PR fix this small change by using "low" instead of "minimal"

ref: https://aistudio.google.com/docs/latest-model#understanding-reasoning-levels

@nmatton

nmatton commented Aug 31, 2026

Copy link
Copy Markdown
Author

This documentation more clearly shows that the "minimal" option was available on 3.6-flash but not anymore on 3.7-flash : https://ai.google.dev/gemini-api/docs/thinking#thinking-levels

using the gemini-latest-flash auto upgraded to 3.7-flash, breaking the app

…lly subject to 503 errors due to high demand.
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.

1 participant