Skip to content

fix(agents): expose GPT reasoning effort levels - #606

Open
Filip-Stastny-Philips wants to merge 4 commits into
codemie-ai:mainfrom
Filip-Stastny-Philips:fix/reasoning-level-support
Open

Filip-Stastny-Philips wants to merge 4 commits into
codemie-ai:mainfrom
Filip-Stastny-Philips:fix/reasoning-level-support

Conversation

@Filip-Stastny-Philips

@Filip-Stastny-Philips Filip-Stastny-Philips commented Oct 3, 2026 •

Copy link
Copy Markdown
Contributor

Summary

Expose model-native reasoning effort levels for GPT-5.6 and GPT-6 in CodeMie OpenCode.

Changes

  • Forward CodeMie effort settings to OpenCode.
  • Add model-specific variants, including xhigh and max.
  • Map minimal to each model’s lowest supported level.

Impact

Selected effort levels are passed through without being downgraded.

@Filip-Stastny-Philips
Filip-Stastny-Philips force-pushed the fix/reasoning-level-support branch from 32b3335 to fcfa1a9 Compare October 7, 2026 22:03
@Filip-Stastny-Philips Filip-Stastny-Philips changed the title fix(agents): preserve extended reasoning effort levels fix(agents): expose GPT reasoning effort levels Oct 7, 2026
@Filip-Stastny-Philips
Filip-Stastny-Philips marked this pull request as ready for review October 7, 2026 22:11

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.

1 participant