Retrieval Augemented Generation (RAG)
A blueprint for RAG on the Hugging Face Hub. https://github.com/huggingface/ai-blueprint
Running4⚡Note Generate responses using: https://huggingface.co/prithivMLmods/SmolLM2-135M-Instruct-GGUF Notebook: https://github.com/huggingface/ai-blueprint/blob/main/rag/generate.ipynb
Running1📚RAG - augment
Note Rerank documents with a query. Notebook: https://github.com/huggingface/ai-blueprint/blob/main/rag/augment.ipynb
Running3🐠RAG - retrieve
Note Vector search on top of Hub datasets by passing a vector and column. Notebook: https://github.com/huggingface/ai-blueprint/blob/main/rag/retrieve.ipynb
ai-blueprint/fineweb-bbc-news-chunked-embeddings
Viewer • Updated • 1.21M • 2Note Chunked and embedded version of https://huggingface.co/datasets/ai-blueprint/fineweb-bbc-news
ai-blueprint/fineweb-bbc-news-embeddings
Viewer • Updated • 353k • 26Note Embedded version of https://huggingface.co/datasets/ai-blueprint/fineweb-bbc-news Notebook: https://github.com/huggingface/ai-blueprint/blob/main/rag/retrieve.ipynb
ai-blueprint/fineweb-bbc-news
Viewer • Updated • 353k • 6Note Your company documents