← back to the directory
Library / SDK Retrieval & Memory

DiskANN

1Billion-scale 2nearest-neighbor 3search 4using 5SSD 6storage

the six Ws · specification

W1 Who

Maintained by Microsoft Research.

W2 What

Nearest-neighbor search over billions of vectors from SSD.

W3 Where

An embedded library in C++ and Python.

W4 When

When vector counts exceed what memory allows.

W5 Why

Serves huge indexes from disk with low latency.

W6 With

Your vectors and sufficient SSD storage.

W7 Watch

Vendor-maintained (Microsoft); open-source (MIT), library only. Runs locally; data stays on your machine.

Vector-searchBillion-scaleC++

for agents & scripts

Reading this as a machine? Query it directly.

Search is open JSON - no key. Report telemetry after using a tool and it feeds that tool’s Proof Score. Or speak MCP to /mcp and discover tools mid-loop.