NVIDIA · Training Benchmark Results
NVIDIA Vera Rubin NVL72 in MLPerf Training v6.1
NVIDIA reports that its Vera Rubin NVL72 platform debuted in MLPerf Training v6.1, with the fastest time to train on six of seven benchmarks and up to 2x per-GPU performance over an optimised GB300 NVL72 baseline. On DeepSeek-V3 671B (a Mixture-of-Experts model), it trained in 4.14 minutes on 1,024 GPUs, against 8.03 minutes for GB300 NVL72, and it kept above 97% scaling efficiency from 256 to 1,024 GPUs. The page says the systems are generally available through cloud infrastructure partners.
Open the source ↗Similar Developer Tools releases
- AWS Emulator
AWS · Agent-built AWS emulator · Week 47, 2026
- Token Pooling-Aware Training for Multi-Vector Embedding Models
Hugging Face (Tom Aarsen) · Pooling-aware embeddings · Week 47, 2026
- WorldFoundry
OpenEnvision (Juanxi Tian, Jiadong Hou, Baiang Li, Xin Lin, Zhengbo Jiao, Weijia Zhang, Yeying Jin, Truong Nguyen) · World model infrastructure · Week 47, 2026
- verifiers
Prime Intellect · LLM RL environments · Week 45, 2026
- SkillSpector v2.11.0
NVIDIA · Agent Skill Scanner · Week 36, 2026
- idea-reality-mcp
mnemox-ai · Idea Reality Checker · Week 36, 2026
More from NVIDIA
- Nemotron RL Litmus-Bench V2.0
NVIDIA · Science Post-Training Dataset · Week 48, 2026
- NVIDIA Rubin HPC
NVIDIA · HPC GPU for Agents · Week 48, 2026
- NVIDIA CUDA-X libraries (cuPhoton, cuPotential, cuSpa, Daviator, rapids-singlecell)
NVIDIA · GPU Science Libraries · Week 48, 2026
- NVIDIA DGX Spark HPC Developer Kit
NVIDIA · Desktop AI Cluster · Week 48, 2026
- cuda.bindings.cudla (cuDLA Python API)
NVIDIA · Python DLA Runtime · Week 47, 2026
- cuBLAS Ozaki-II and 2M FP64 Algorithms
NVIDIA · FP64 GEMM Emulation · Week 47, 2026
Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.