{"repo":"Egoist-Machines/LodeDB","free":true,"listed":false,"github":"https://github.com/Egoist-Machines/LodeDB","clone":"git clone https://github.com/Egoist-Machines/LodeDB.git","description":"World's fastest and most compact embedded vector database: exact by default, multimodal, local-first, and GPU-accelerated","language":"Python","stars":95,"topics":["db","embeddings","gpu-acceleration","local-first","privacy","rag","semantic-search","turbovec","vector-database","clip"],"license":"Apache-2.0","category":"databases-storage","readme_excerpt":"LodeDB 🔥 the fastest and most compact embedded vector database in the world 🌍 Built by Egoist Machines, Inc. - efficient full-stack infrastructure for reliable AI systems. LodeDB is great for local RAG; it's extremely fast , exact by default, in-process, and on-disk. We're the best drop-in durable memory backend for LangChain, LlamaIndex, and mem0 : the most compact on disk, the fastest per single query, GPU-accelerated for batched search, and durable in about a millisecond per write. Point any of them at LodeDB instead of its default store. Over 17.5k documents, per framework default: vs the framework's default store LangChain InMemoryVectorStore LlamaIndex SimpleVectorStore mem0 Qdrant --- --- --- --- On-disk footprint 13.6× smaller (15 vs 199 MB) 9.9× smaller (15 vs 145 MB) 5.7× smaller (12 vs 70 MB) Single-query p50 (CPU) 600× faster (0.45 vs 272 ms) 620× faster (0.44 vs 272 ms) 46× faster (0.59 vs 27 ms) Batched retrieval, 64 (GPU) 2,880× (11,049 vs 4 qps) 3,050× (11,297 vs 4 qps) 139× (5,084 vs 36 qps) Durable add of one memory 26,000× faster (0.26 ms vs 6.9 s) 57,000× faster (0.26 ms vs 14.8 s) 0.28 vs 0.44 ms (both sub-ms) Among embedded stores, LodeDB has the smallest footprint and the fastest single-query and batched search, and its durable add leads the fastest lazy-append stores (sqlite-vec, qdrant) too: embedded stores durable add p50 single-query p50 batch-64/query memory footprint --- ---: ---: ---: ---: LodeDB 0.26 ms 0.45 ms 0.09 ms 15 MB sqlite-vec 0.42 ms","default_branch":null,"files":null,"tree":[],"storefront":"/r/Egoist-Machines","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/Egoist-Machines/LodeDB/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}