Qdrant

@qdrant_engine

High-performance Rust-based vector search engine. discord.com/invite/qdrant

Qdrant
@qdrant_engine
21h
On Chroma Cloud and looking at Qdrant? Writes stop Nov 22. Book a call with one of our engineers and plan the cutover together: which collections go first, how to check results match, what changes in your query code. qdrant.tech/migrate-from-c…
Qdrant
@qdrant_engine
21h
One week until our meetup in NYC. RSVP for debates, drinks, and prizes. November 10 | 6:30 PM ET | 87 Lafayette St | with @AgoraIO luma.com/nyc-meetup-nov…
Qdrant
@qdrant_engine
Nov 2
Same query. 10× higher latency. In our 4.6M-vector test, rescoring rose from 4.1ms to 43.4ms when RAM dropped from 12 to 4 GiB. The culprit? Reading original vectors from disk. Why memory limits matter: qdrant.tech/articles/when-…
3
Qdrant
@qdrant_engine
Oct 29
Join Qdrant and @AgoraIO on November 10 for an AI debate night. Pick one of three takes, including "RAG is dead." Get your side at random, sell it in 10 minutes, and the room votes. Or simply spectate, up to you. Winners take home a Nintendo Switch Lite, a Lego Technic NASA
1
Qdrant
@qdrant_engine
Oct 29
780 ms → 10.2 ms median search latency during optimization. We tested 13 Qdrant optimizer configurations on 1.76M vectors. prevent_unoptimized helped, but delays visibility of new points until indexing finishes. qdrant.tech/articles/tunin…
8