Nebius GB300 NVL72 rack tops MLPerf with 603k tokens/sec on DeepSeek R1
demian_ai · x · 2026-09-18
Nebius claimed five first-place results in MLPerf Inference v6.1: a full 72-GPU GB300 NVL72 rack topped DeepSeek R1 in both server and offline scenarios (603k and 690k tokens/sec) and pushed gpt-oss 120B past 1M tokens/sec. Nebius was one of only two submitters with preview results on NVIDIA's Vera Rubin NVL72. Scaling from 8 to 72 GPUs delivered near-linear 9x speedup.
More from Infra
- Simulating 200,000 months: can your home AI rig pay rent at 2026 GPU prices? — LeviTurk · 2026-11-02
- Tailscale called a 'single point of failure' for startups as acquisition chatter splits devs — TheZachMueller · 2026-11-02
- Gary Marcus: AI compute is headed for a glut, just like fiber in the dot-com bust — GaryMarcus · 2026-11-02
- One H200 Isn't Enough for DeepSeek 4.1 Flash: Two GPUs at $10.59/hr — daluoseo · 2026-11-02
- Databases to distributed systems: every systems skill is merging into AI inference and hardware — Abhishekcur · 2026-11-02
- AWS not "sold out" until 2027: expert says capacity headline needs context — DavidLinthicum · 2026-11-02