Qdrant benchmarks 13 optimizer configs on 1.76M vectors: 780ms vs 4.3ms search latency
qdrant_engine · x · 2026-10-13
Qdrant published a benchmark of 13 optimizer configurations on 1.76M Cohere-embedded MS MARCO vectors (single node, 32GB RAM, 14 cores) to quantify trade-offs between indexing speed, recovery time, and query latency.
- Search latency hit 780ms while background optimizers were running, dropping to 4.3ms once settled — a 200x gap on the same collection.
- Each run measured three stages: Upload (70–316s depending on concurrent indexing), Draining (p50/p95 while indexing/merge/vacuum work through the backlog), and Steady baseline (5,000 fixed queries).
- Client ran on the same machine to eliminate network latency; cloud results would likely be worse.
Related event: Qdrant Benchmarks Show Optimizer Config Drastically Affects Search Latency(2 posts)→
More from Infra
- Yoniq Compute repo ships inference recipes validated on 1x-8x RTX Pro 6000 with SGLang and vLLM — TheZachMueller · 2026-10-13
- RL post-training compute accounting: one weight update per episode blurs the train/inference line — davidmanheim · 2026-10-13
- Ex-Apple engineer eyeing Surface with NVIDIA RTX Spark: 128GB + CUDA in a backpack — LinusEkenstam · 2026-10-13
- Local 125B MoE on RTX 5090 finishes Three.js horse challenge in 23 minutes — Beneficial-Ad-8127 · 2026-10-13
- ML Veteran John Langford Endorses GPU MODE's Open Benchmark Competition Model — JohnCLangford · 2026-10-13
- Transcribing an hour of audio in ~13s on a MacBook Air — amrrs · 2026-10-13