NVIDIA Wins Every MLPerf Training v5.1 Benchmark
Follow @NVIDIANewsroom for real-time updates from NVIDIA.
NVIDIA Wins Every MLPerf Training v5.1 Benchmark.
In the age of AI reasoning, training smarter, more capable models is critical to scaling intelligence. Delivering the massive performance to meet this new age requires breakthroughs across GPUs, CPUs, NICs, scale-up and scale-out networking, system architectures, and mountains of software and algorithms.
In MLPerf Training v5.1 — the latest round in a long-running series of industry-standard tests of AI training performance — NVIDIA swept all seven tests, delivering the fastest time to train across large language models (LLMs), image generation, recommender systems, computer vision and graph neural networks.
NVIDIA was also the only platform to submit results on every test, underscoring the rich programmability of NVIDIA GPUs, and the maturity and versatility of its CUDA software stack.
The GB300 NVL72 rack-scale system, powered by the NVIDIA Blackwell Ultra GPU architecture, made its debut in MLPerf Training this round, following a record-setting showing in the most recent MLPerf Inference round.
Compared with the prior-generation Hopper architecture, the Blackwell Ultra-based GB300 NVL72 delivered more than 4x the Llama 3.1 405B pretraining and nearly 5x the Llama 2 70B LoRA fine-tuning performance using the same number of GPUs.
Mayor Matt Mahan on the NVIDIA AI Podcast: How AI is Changing City Life in San Jose.
Mayor Matt Mahan and NVIDIA’s Jumbi Edulbehram reveal how AI is making San Jose smarter — optimizing transit, upskilling city staff, and powering pioneering civic programs, and more.
Learn how AI is helping streamline services and create a more responsive city for everyone — listen to the full podcast.
ICYMI: America’s Next Apollo Moment is Unveiled at GTC in Washington D.C.
From the transistor to the internet, every leap forward has shaped America’s destiny. And now, a new industrial revolution begins — powered by AI and NVIDIA accelerated computing.
This is America’s next Apollo moment. The age of AI.
In case you missed it, watch NVIDIA CEO Jensen Huang tell the story at his GTC keynote in Washington D.C.
From Our Leaders
In The News
The new Fairwater AI datacenter in Atlanta, powered by hundreds of thousands NVIDIA Blackwell and Blackwell Ultra GPUs leveraging the NVIDIA GB200 and GB300 NVL72 rack-scale systems with Spectrum-X Ethernet, is pushing boundaries for performance, sustainability, and AI infrastructure.
NVIDIA is proud to be a part of the next chapter in AI innovation alongside Microsoft.
What's Ahead
Discover how the intersection of supercomputing and science is being redefined. Join our NVIDIA SC25 Fireside Chat to hear directly from the minds shaping the future of HPC.
📌 Reserve your spot: https://nvda.ws/47nWdWh
Stay Connected
The innovation doesn’t stop here. Stay informed with these resources from NVIDIA:
- 📰 Read the NVIDIA Newsroom blog
- 📱 Follow @NVIDIANewsroom X, Instagram
- 🏛️ Making an Impact: AI for Public Policy
President, Front Page News Entertainment Group Worldwide I grew my company and now accelerating in Media, Entertainment, Music Distribution, Sales, Advising Professional Athletes, Brands and Building Companies.
1dYes to performance tests and benchmarks!
President, Front Page News Entertainment Group Worldwide I grew my company and now accelerating in Media, Entertainment, Music Distribution, Sales, Advising Professional Athletes, Brands and Building Companies.
1dThank you for sharing
Senior Technical Program Manager @ CMIT Solutions | Cloud Applications, Digital Transformation
4d👍
Sr. Technical Program Manager @ Intel • AI/ML & GenAI • Semiconductor & GPU QA • Data Center & Cloud Infrastructure • Quality Engineering •SRE • SAP S/4HANA Test Lead • Enterprise Program Delivery (Google, Meta, Amazon)
4dImpressive milestones!