Nvidia (NVDA) says Vera Rubin trains AI up to 2x faster. The 2x came on one of seven tests.

Nvidia logo above MLPerf training times on DeepSeek-V3 with 1,024 GPUs: Vera Rubin 4.14 minutes, GB300 Blackwell Ultra 8.03 minutes

Key points

  • Nvidia's Rubin makes its MLPerf training debut
  • CoreWeave tests Rubin across four racks
  • Nvidia's Kimi K3 result awaits verification

Nvidia (NVDA) says its new Vera Rubin NVL72 system delivered up to twice the per-GPU training performance of Blackwell Ultra in its first MLPerf training round. The biggest gain came on a DeepSeek-V3 test, where the two systems each ran on 1,024 GPUs.

Rubin posted the round's fastest training time on six of seven tests, according to Nvidia. The exception was DeepSeek-V3, where a much larger system using Nvidia's older GB300 chips finished first.

The industry group MLCommons published the Training v6.1 results on Nov. 16, a day before Nvidia reports quarterly earnings. Twenty-four organizations submitted 153 systems. The round included Nvidia's Rubin GPUs and Google's TPUv7, also called Ironwood, for the first time.

Where Nvidia's 2x claim comes from

MLPerf's training tests measure how long a system takes to train a model to a set quality target. On DeepSeek-V3, a 671-billion-parameter model, a Vera Rubin system with 1,024 GPUs finished in 4.14 minutes. A GB300 NVL72 system with the same number of GPUs took 8.03 minutes.

Nvidia calls that its largest per-GPU gain over GB300, its Blackwell Ultra system. The company puts Rubin's gain on the other six tests at 1.4x to 1.6x.

Nvidia also ran DeepSeek-V3 on 256 Rubin GPUs. Increasing the count fourfold to 1,024 cut the training time from 16 minutes to about four. Nvidia puts the result at about 97% scaling efficiency. Perfect scaling would reduce the time to one-quarter of the original.

The company published faster reruns using newer software, too. Nvidia says those reruns haven't been verified by MLCommons and aren't part of the official results.

A larger GB300 system finished DeepSeek-V3 first

Rubin's lead over GB300 on DeepSeek-V3 compares systems with the same number of GPUs. The fastest DeepSeek-V3 time in the round came from a much larger Nvidia GB300 system: 10,240 GPUs finished in 1.38 minutes. Microsoft (MSFT) also ran the test on 10,240 GPUs in its Azure cloud, using GB200 systems, the generation before GB300.

Across its platforms, Nvidia says its systems held the fastest time on all seven tests and were the only ones entered in all seven. AMD (AMD) entered five tests, according to its folder in the MLCommons results repository. Google (GOOGL) entered TPUv7 in two: DeepSeek-V3 and a Qwen3.5 reinforcement learning test.

CoreWeave ran Rubin across four racks

CoreWeave (CRWV) said it was the only cloud provider to submit multi-rack Vera Rubin results in the round.

CoreWeave's 256 Rubin GPUs, spread across four racks, trained DeepSeek-V3 in 17.65 minutes. CoreWeave's GB200 NVL72 systems took 48.40 minutes with the same number of GPUs, which the company calls a 2.7x improvement.

CoreWeave also ran Llama 3.1 8B on a single 72-GPU rack, finishing in 7.16 minutes. The company says that was the fastest 72-GPU submission in the round. Each Vera Rubin NVL72 rack pairs 72 Rubin GPUs with 36 of Nvidia's Vera CPUs.

Nvidia's 10x claim uses a different benchmark

Nvidia also says Vera Rubin delivered more than 10 times the throughput per watt of GB300 on Kimi K3, an open model built for agent workflows. That claim comes from a different benchmark, MLPerf Endpoints v1.0, which tests systems running AI agents on coding and business tasks. The results were published on Nov. 9.

The Rubin result is provisional. Nvidia says it is pending MLCommons verification and could change after peer review.

The comparison used 32 accelerators in the Rubin system and 48 in the GB300 system. Nvidia measured output per kilowatt at a minimum response speed of 50 tokens per second per user.

Nvidia reports earnings on Nov. 17

The results arrive ahead of Nvidia's fiscal third-quarter report, due Nov. 17 at about 4:20 p.m. ET. It will be the company's first earnings report since Jensen Huang resigned as CEO.

The Rubin-versus-GB300 and Rubin-versus-GB200 speedups are Nvidia's and CoreWeave's comparisons of results published by MLCommons.

Frequently asked questions

What were Nvidia Vera Rubin's MLPerf Training results?

Nvidia says its Vera Rubin NVL72 system posted the fastest time to train on six of the seven tests in MLPerf Training v6.1, which MLCommons published on November 16, 2026. On DeepSeek-V3, it trained the model in 4.14 minutes on 1,024 GPUs, against 8.03 minutes for Nvidia's GB300 NVL72 on the same number of GPUs.

Is Nvidia Vera Rubin 2x faster than Blackwell Ultra?

Only on one test. Nvidia's up-to-2x per-GPU figure comes from the DeepSeek-V3 benchmark in MLPerf Training v6.1. Nvidia puts Vera Rubin's per-GPU gain over its Blackwell Ultra system, GB300 NVL72, at 1.4x to 1.6x on the other six tests.

Did Nvidia win every MLPerf Training v6.1 benchmark?

Nvidia says its platforms together hold the fastest time on all seven tests, but Vera Rubin led six. The fastest DeepSeek-V3 time, 1.38 minutes, came from Nvidia's older GB300 systems on 10,240 GPUs. AMD entered five tests and Google entered its TPUv7 chips in two.

Where does Nvidia's 10x Kimi K3 claim come from?

It comes from MLPerf Endpoints v1.0, an MLCommons test of AI agent workloads published on November 9, 2026, not from the training round. Nvidia says Vera Rubin NVL72 delivered more than 10 times the throughput per watt of GB300 NVL72 on Kimi K3. The Rubin result is provisional and pending MLCommons verification.

More on AMD and CRWV

Latest stories

Dennis Singleton
Dennis Singleton

Dennis Singleton was born in Australia and later moved to the United States. He has spent years following the markets, but what keeps his attention is how AI is built. He writes about the companies behind the technology, from semiconductor designers and advanced packaging to photonics, memory, networking, and the hardware powering modern AI. His approach starts with filings, earnings, and industry research, then translates the important details into clear, straightforward analysis without unnecessary hype.