Skip to content

NVIDIA Vera Rubin Debuts With Leading MLPerf Training and Agentic Inference Results

NVIDIA BlogFrom the source· Shruti Koparkar·· 9 hours agoPicksNVIDIA Sweeps MLPerf Training and Endpoints as Vera Rubin Debuts With Leading Performance
Original
Image: NVIDIA Blog
Who
NVIDIA
What
reported benchmark results Vera Rubin platform results in MLPerf Training v6.1 and MLPerf Endpoints v1.0
When
Quote
In its MLPerf benchmarks debut, with results released today, the NVIDIA Vera Rubin platform delivers leading performance across training, post-training and agentic inference.

What happened

NVIDIA says its Vera Rubin platform swept every MLPerf Training v6.1 benchmark and led across MLPerf Endpoints agentic benchmarks in its debut. On DeepSeek-V3 training, it scaled from 256 to 1,024 GPUs with about 97% efficiency; a provisional Kimi K3 result reports more than 10x the throughput per watt of GB300 NVL72 at a shared minimum interactivity of 50 tokens per second per user. The Vera Rubin Endpoints result is pending MLCommons verification.

Why it matters

Vera Rubin scaled DeepSeek-V3 training from 256 to 1,024 GPUs with about 97% efficiency, while its provisional Kimi K3 result reports over 10x the throughput per watt of GB300 at a shared response speed.

Read next

See the full storyNVIDIA Reports Vera Rubin Debut Results in MLPerf Training and Endpoints

The story so far

NVIDIA said its Vera Rubin platform swept every MLPerf Training v6.1 benchmark and led across MLPerf Endpoints agentic benchmarks in its debut, though its Endpoints results remain pending MLCommons Association verification. On DeepSeek-V3 training, NVIDIA reported that Vera Rubin scaled from 256 to 1,024 GPUs with about 97% efficiency. For MLPerf Endpoints v1.0, a provisional Kimi K3 result showed more than 10x the throughput per watt of GB300 NVL72 at a shared minimum interactivity of 50 tokens per second per user. NVIDIA noted that the provisional Vera Rubin results are subject to change pending peer review.

This story is in 11/17 AI DailyRead the whole issue