Skip to content

Release v0.1.16-hotfix1

Choose a tag to compare

View all tags
@github-actions github-actions released this 04 Aug 22:21
v0.1.16-hotfix1
61ab38a

What's Changed

  • this hotfix just fixes a debug flag that was inadvetently left hardcoded (logging)

  • Custom Prompts: Customize every aspect of how LLxprt prompts your model, over-ride the defaults on a per-model basis!

  • No nextSpeakerChecker - removed gemini-cli artifact that sends the entire conversation to flash multiple times per request to see if Gemini-2.5-pro should continue talking.

  • Security improvements: /key now *** masks the input

  • Provider Configurations: /profile save "myConfig" /profile load "myConfig" --profile-load - you can now save your whole provider/model/settings to a profile and use it on startup or in the non-intractive mode.

  • Ephemeral Settings: /set context-limit compression-threshold etc.

  • Model params: /set modelparam max_tokens 100000 /set modelparam Enable_Thinking True

  • Multibyte character support (incl on windows) fixes

  • Fix accidental flash fallback revert

Full Changelog: v0.1.15...v0.1.16

Assets 3
Loading