MLPerf Storage v3.0 lands with 144 results, adding KV cache and vector DB tests
TheKanter · x · 2026-09-02
MLCommons released MLPerf Storage v3.0 results: 144 results from 19 organizations, 11 first-time submitters. The suite now covers four workloads — training, checkpointing, vector database, and KV cache — with checkpointing tested from 8B up to 1250B parameters.
Key additions:
- A KV Cache test measuring storage performance for LLM inference cache read/write;
- A Vector DB test for high-dimensional indexing and query workloads;
- S3 object storage API support alongside POSIX;
- New efficiency metrics: rack-unit density and provisioned power.
Working group co-chair Brian Belgodere said decomposing monolithic AI systems into storage patterns like checkpointing, KV caching, and vector DBs gives stakeholders a much clearer view of AI storage needs.
More from Infra
- Hardware Advice: Running H3 Mini Max Locally on Mac — KourriX · 2026-09-02
- PIXIO Report: Doubling the speed of open video model inference — tsi_org · 2026-09-02
- AMD MI100 User Report: Cooling Struggles and Subpar Inference Speeds — faisalkl · 2026-09-02
- Ramp builds a semantic layer linking AI spend to ROI by decoding agent traces — graceisford · 2026-09-02
- Analyst: Dell's supply chain signals are 'ground truth' for AI demand — BenBajarin · 2026-09-02
- Strix Halo + 3090 Ti Pushes Qwen Flash-Next to 84 tok/s via Deep Optimization — TrifleHopeful5418 · 2026-09-02