v3.7.0
📦 gpt4all
✨ 3 features🐛 5 fixes
Summary
This release introduces official support for Windows ARM devices and includes significant stability fixes for macOS update/quit operations, alongside improvements to the Code Interpreter and chat template parsing.
Migration Steps
- If you previously installed the GitHub release version of GPT4All as a workaround for macOS update issues, you can safely uninstall it and switch back to the version downloaded from the official website.
- Windows ARM users must install the new specific *Windows ARM* version of GPT4All from the website; the standard *Windows* version will not work.
✨ New Features
- Added support for the Windows ARM platform (Qualcomm Snapdragon and Microsoft SQ-series processors).
- Improved behavior when the Code Interpreter takes too long to execute and times out.
- Added support for multiple arguments in console.log within the Code Interpreter for better JavaScript compatibility.
🐛 Bug Fixes
- Fixed maintenance tool crashing when updating or uninstalling GPT4All on macOS Sequoia.
- Fixed chats not saving correctly when the application is quit using Command-Q on macOS.
- Fixed two crashes and one compatibility issue in the chat template parser.
- Fixed the default chat template for EM German Mistral.
- Added automatic replacements for five new models to improve chat template compatibility.