v1.11

1. Fixed model thinking mode auto-detection bug, added auto-detection support for inkling-small, minimax-m3, and deepseek-v4-flash-0731 models
2. Fixed model parameter import bug and AI chat page code rendering issues
3. AI Chat page UI display optimization

v1.1

1. Ultimate Edition AI chat supports automatic detection of local models, no manual API configuration needed; thinking level can be selected at model startup or during conversation
2. Added English language support, with Chinese/English switching
3. Added more llama.cpp parameter support, including primary GPU, repeat-penalty, top-k, top-p, etc.
4. Fixed various bugs, including AI chat window abnormal closure prompts, path search popup height issues, etc.

v1.0

First release version of hybridLLM.

For full features, purchase guides, and changelogs, please refer to the main repository README.