๐๐น๐๐ฎ๐๐ ๐๐ฎ๐ป๐๐ฒ๐ฑ ๐๐ผ ๐ฐ๐ต๐ฎ๐ ๐๐ถ๐๐ต ๐๐ผ๐๐ฟ ๐ฝ๐ฑ๐ณ๐ ๐ถ๐ป ๐ฎ ๐๐ถ๐บ๐ฝ๐น๐ฒ ๐ฎ๐ป๐ฑ ๐ฒ๐ฎ๐๐ถ๐น๐-๐ฎ๐ฐ๐ฐ๐ฒ๐๐๐ถ๐ฏ๐น๐ฒ ๐๐ฎ๐?
Now you can! I am thrilled to introduce ๐ฒ๐๐ฒ๐ฟ๐๐๐ต๐ถ๐ป๐ด-๐ฟ๐ฎ๐ด, your fully customizable and local chatbot assistant! ๐ค
๐ง๐;๐๐ฅ:
You do not want to read this post but you want to start getting your hands dirty?
Check out the online HuggingFace space built for everything-rag!
With everything-rag, you can:
- Use virtually any LLM you want
- Use your own data: everything-rag can work with any pdf you provide, whether it's about data sciences or pallas' cats!๐
- Enjoy 100% local and 100% free functionality: no hosted APIs or pay-as-you-go services.
- Exploit Retrieval Augmented Generation
- Simple installation through Docker image
This is completely free to use and runs on your local computer!
Make sure to check out (and leave a little โญ if you please) the GitHub repo, have a look to my blog post about the importance of open-source LLMs and, also, do not forget to sponsor me on GitHub if you please!
Inspired by Jan.ai (for local LLM development) and Cheshire Cat AI (for the elegant dockerized installation workflow).
Shout-outs to Hugging Face (LLM support and Online space), Gradio (app interface), Langchain (retrieval architecture), Chroma (vectorized databases architecture), Docker and GitHub (docker image distribution and hosting) for having made everything-rag possible!
And stay tuned, more features are coming in the future!๐