RealityHackerOpen in RealityHacker ⇢
Models · Fine-Tuning & Adaptation · published 2026-10-27T00:00:00+00:00 · via MarkTechPost

UC Berkeley Unveils LEANN: A Compact Open-Source Vector Index for Efficient Retrieval

LEANN, developed by UC Berkeley, reduces the storage footprint of large-scale vector indexes by leveraging a pruned graph and on-the-fly embedding recomputation. It achieves comparable accuracy to HNSW at 90% recall while using only 4 GB of memory instead of 188 GB. The trade-off is slightly longer query times, measured in seconds rather than milliseconds.

Read the full article at MarkTechPost →
This summary is Al-enhanced to contain extended analysis and broader social context. The original is {NAME); the linked article is the authoritative source. Original headline: “Meet LEANN: An Open-Source Vector Index that Shrinks a 60M-Chunk RAG Index from 201 GB to 6 GB.” Browse more stories.