[MLsys2026 Best Paper]: https://arxiv.org/abs/2506.08276. RAG on Everything with LEANN. Enjoy 97% storage savings while running a fast, accurate, and 100% private RAG application on your personal device.
Python aifaissgpt-osslangchainllama-indexllmlocalstorageoffline-firstollamaprivacypythonragretrieval-augmented-generationvector-databasevector-searchvectors
merge main
yichuan520030910320df63526503b0
diff --cc examples/main_cli_example.pyindex 0439015,1cfb4d2..d489970--- a/examples/main_cli_example.py+++ b/examples/main_cli_example.py@@@ -68,9 -73,9 +73,8 @@@ async def main()chat = LeannChat(index_path=INDEX_PATH)query = "Based on the paper, what are the main techniques LEANN explores to reduce the storage overhead and DLPM explore to achieve Fairness and Efiiciency trade-off?"- # query = "What is the Off-policy training in RL?"print(f"You: {query}")-- chat_response = chat.ask(query, top_k=20)+ chat_response = chat.ask(query, top_k=20, recompute_beighbor_embeddings=True,embedding_model="facebook/contriever")print(f"Leann: {chat_response}")if __name__ == "__main__":