v3.10.0
📦 gpt4allView on GitHub →
✨ 6 features🐛 1 fixes🔧 3 symbols
Summary
This release introduces dedicated configuration tabs for remote models like Groq, OpenAI, and Mistral, and improves compatibility by supporting older CUDA devices. It also includes fixes for crashes and better handling of whitespace for DeepSeek-R1 models.
✨ New Features
- Dedicated tab for remote model providers on the Add Model page.
- Easier configuration for Groq, OpenAI, and Mistral remote models.
- Support added for the non-MoE Granite model.
- Improved Italian and significantly improved Simplified Chinese translations.
- Improved default chat templates for OLMoE 7B 0924/0125 and Granite 3.1 3B/8B.
- Better whitespace behavior in output for DeepSeek-R1-based models.
🐛 Bug Fixes
- Fixed several issues that could potentially cause GPT4All to crash.
🔧 Affected Symbols
DeepSeek-R1-based modelsOLMoE 7B 0924/0125Granite 3.1 3B/8B