Feature Updates
- Consolidated offline English documentation #39
- Support for direct integration with third-party APIs. The default model now includes Groq Llama3.1 70B.
- A Pre-built Docker Image is provided for faster installation.
- The Docker version directly supports CUDA, eliminating the need for repeated container installation.
- Added Kuwa Javascript Library for calling Multi-chat API.
- Ability to directly edit Bot settings by clicking the Bot image within the chat room.
- RAG supports caching Embedding models and Vector DB.
- SearchQA facilitates integration with third-party search engines.
- DocQA/WebQA offers a Failback mode, allowing direct interaction with an LLM.
- Pipes can now specify parameters from the prompt.
- Added Media Converter for editing and splicing videos or audio.
- The default Gemini model has been updated to Gemini 1.5 Flash.
Bug Fixes
- Docker WARN message "FromAsCasing: 'as' and 'FROM' keywords' casing do not match" #38
- The Docker Version Executor diagram disappears
- Chinese display garbled on the Windows version
- The Chinese file path is not supported when creating Vector DB on the Windows version
- The file name is too long and exceeds the path length limit on the Windows version
- ChatGPT executor does not follow custom context_window
Tutorial
info
- Kuwa-v0.3.4 download information: https://github.com/kuwaai/genai-os/releases/tag/v0.3.4
- Kuwa-v0.3.4 single executable download link: https://github.com/kuwaai/genai-os/releases/download/v0.3.4/Kuwa-GenAI-OS-v0.3.4.exe
warning
The Docker version defaults to using pre-build Docker images, if you need to build from the source, please refer to this document.
We welcome your feedback after experiencing it. If you encounter any difficulties, please feel free to contact us through community or other channels.
Kuwa official website: https://kuwaai.tw/
Kuwa introduction: https://kuwaai.tw/blog/kuwa-os-intro