[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
upd readme
yichuan5200309103202a96d05b21a7
diff --git a/README.md b/README.mdindex 3624c81..8f62ca8 100755--- a/README.md+++ b/README.md@@ -389,15 +389,7 @@ python examples/compare_faiss_vs_leann.pySame dataset, same hardware, same embedding model. LEANN just works better.-## Reproduce Our Results--```bash-uv pip install -e ".[dev]" # Install dev dependencies-python examples/run_evaluation.py data/indices/dpr/dpr_diskann # DPR dataset-python examples/run_evaluation.py data/indices/rpj_wiki/rpj_wiki.index # Wikipedia-```-The evaluation script downloads data automatically on first run.### Storage Usage Comparison@@ -424,6 +416,15 @@ The evaluation script downloads data automatically on first run.*Benchmarks run on Apple M3 Pro 36 GB*+## Reproduce Our Results++```bash+uv pip install -e ".[dev]" # Install dev dependencies+python examples/run_evaluation.py data/indices/dpr/dpr_diskann # DPR dataset+python examples/run_evaluation.py data/indices/rpj_wiki/rpj_wiki.index # Wikipedia+```++The evaluation script downloads data automatically on first run. The last three results were tested with partial personal data, and you can reproduce them with your own data!## 🔬 PaperIf you find Leann useful, please cite:diff --git a/packages/leann-backend-diskann/third_party/DiskANN b/packages/leann-backend-diskann/third_party/DiskANNindex 25339b0..af2a264 160000--- a/packages/leann-backend-diskann/third_party/DiskANN+++ b/packages/leann-backend-diskann/third_party/DiskANN@@ -1 +1 @@-Subproject commit 25339b03413b5067c25b6092ea3e0f77ef8515c8+Subproject commit af2a26481e65232b57b82d96e68833cdee9f7635