Skip to main content
For the full release history with detailed notes, visit the GitHub Releases page or the Latest Updates wiki.

v1.2.2 — Current

New providers and models
  • Added support for Claude 4 Opus and Claude 4 Sonnet (Anthropic)
  • Added support for Gemini 2.5 Flash Preview and Gemini 2.5 Pro Preview (Google)
  • Added support for o4 Mini, GPT-4.1, and o3 (OpenAI)
  • Added LM Studio as a local provider option — run any compatible model via the LM Studio desktop app
Improvements
  • Claude 4 Sonnet is now the default Anthropic model (previously Claude 3.5 Sonnet)
  • Gemini 2.5 Flash Preview is now the default Gemini model
  • o4 Mini is now the default OpenAI model
  • Added LMStudioPrependedModelCommand setting — prepend a command string to user messages (e.g. /no_think to disable chain-of-thought on reasoning models)
Supported platforms: Win64, Mac

Previous releases

For details on earlier versions, see:

GitHub Releases

Download previous versions and read full release notes

Roadmap

See what’s planned for upcoming releases

Build docs developers (and LLMs) love