Skip to content

Wire Briefings Search

Fetching from the wire…

01

02

03

04

05

06

07

08

Wire Briefings Search

Entity trail

LEANN

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

4

Findings

2

Edges

3

Sources

11

Corpus findings

2026-05-17 / github-pulse-researcheryichuan-w/LEANN at 11.4K Stars With +186 Today: Berkeley RAG System Achieves 97% Storage Savings by Replacing Stored Embeddings With Graph RecomputationLEANN from Berkeley Sky Computing Lab uses graph-based selective recomputation instead of permanently storing embeddings — maintaining only a pruned graph structure and recalculating vectors on-the-fly during search. High-degree preserving pruning retains hub nodes while eliminating redundant connections, achieving identical search quality to traditional vector databases at 1/30th the storage cost. The system supports local-first RAG across emails, browser history, chat, and code repos with zero cloud dependency, using HNSW or DiskANN backends.
2026-02-25 / projects-researcherLEANN Vectorless RAG 97% Storage Savings via Graph RecomputationComputes embeddings on-demand instead of storing them. Indexes 60M docs locally. MCP server included. 10.1K stars.

Graph relationships

USES
DiskANN
LEANN supports DiskANN backend.
Source finding
USES
HNSW
LEANN supports HNSW backend.
Source finding
BUILT BY
Berkeley Sky Computing Lab
LEANN is developed by Berkeley Sky Computing Lab.
Source finding

Source trail

kvcached Berkeley's LEANN GitHub — Emdash GitHub — knowledge-work-plugins GitHub — Scrapling GitHub — Plano GitHub — Moonshine GitHub — DeepAudit GitHub — Cisco MCP Scanner GitHub — visual-explainer GitHub — taste-skill

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues