NVIDIA's Blackwell platform has achieved top performance across all seven benchmarks in the MLPerf Training 6.0 industry standard tests. The platform demonstrated the fastest training times and enabled the largest-scale training runs, utilizing up to 8,192 GPUs. This success highlights the platform's ability to accelerate AI model development and reduce training costs through advanced hardware and networking capabilities. AI
IMPACT Sets a new standard for AI training infrastructure performance and scale, potentially accelerating the development of larger and more complex AI models.
RANK_REASON The cluster reports on performance results from industry-standard benchmarks (MLPerf Training 6.0) for AI training infrastructure, which falls under research and development achievements.
Read on Mastodon — fosstodon.org →
- Microsoft Azure
- CoreWeave
- DeepSeek-V3 671B
- GPT-OSS-20B
- Llama 3.1 405B
- MLPerf Training 6.0
- NVIDIA
- NVIDIA Blackwell
- NVIDIA GB200 NVL72
- NVIDIA GB300 NVL72
- NVIDIA Nemotron 3 Ultra
- Blackwell
- GB200 NVL72
- GB300 NVL72
- NVFP4
- NVIDIA Blackwell NVL72
- NVLink
AI-generated summary · Google Gemini · from 3 sources. How we write summaries →