Weaviate cut vector index memory from 9.9 GB to 1.4 GB over three years of quantization
CShorten30 · x · 2026-10-23
A slide from Weaviate's all-hands shows memory usage for a 1M-vector DBPedia index (ada-002, 1536-dim, HNSW, k=10) falling from 9.9 GB in v1.23.0 (uncompressed, 2023) to 5.1 GB with scalar quantization (2024), 2.1 GB with rq8 (2025), and 1.4 GB with rq4c (2026) — roughly a 7x reduction driven by successive quantization upgrades.
More from Infra
- One researcher pretrains Mirendil-Base coding model in 3.5 months for under $10M — vjkaruna · 2026-10-23
- Video Model API Markups Exposed: Runway Charges 414%, New Sume API Launches With 0% Markup — Scobleizer · 2026-10-23
- Google Cloud launches Agent Compute Engine, a new VM type built for AI agents — testingcatalog · 2026-10-23
- LM Studio ships Read Aloud, powered by tiny local TTS model Phonon from Gradium — mattturck · 2026-10-23
- Lam Research system revenue up 35% QoQ as AI pushes 2027 WFE past $200B — BenBajarin · 2026-10-23
- Hugging Face Hub launches OpenUSD support with 2,500+ NVIDIA SimReady assets — mishig25 · 2026-10-23