Candidate depth 10→500 lifts score by just 0.01: Qdrant on diagnosing before tuning vector search
qdrant_engine · x · 2026-09-08
Qdrant's team argues that vector search can "work" while still being badly tuned — with knobs like hnswef, candidate depth, RRF k, quantization, and reranking, it's unclear which to change first. Testing across five public datasets, they found the right knob depends on where the problem is:
- Raising candidate depth 10 → 500 improved the best achievable score by up to 0.28, but the final score improved at most 0.01 when relevant docs were already retrieved but ranked too low
- Lesson: diagnose the failure mode first (recall vs ranking), then tune the knob tied to that problem
Their new article lays out a systematic approach to retrieval tuning without guessing.
More from coding & agent
- GPT-6 Astra computer use isn't in Chat — the real surfaces are Work, Codex and desktop — socialwithaayan · 2026-09-08
- Agents hijacking your desktop? Devs say run them in a VM or a dusty Mac mini — paw_lean · 2026-09-08
- Universe: a Mac agent workspace that reuses your Claude or ChatGPT subscription, free then $19/mo — tedddyoweh · 2026-09-08
- Hooking an old telescope to an AI-built dashboard, with a simulator as backup — lvwerra · 2026-09-08
- The unmeasured LLM skill: knowing when to stop and ask instead of guessing — sunychoudhary · 2026-09-08
- Open-source AI skill clears 135GB of Mac cache overnight, saving a ¥1500 SSD upgrade — oran_ge · 2026-09-08