--- title: NVIDIA Blackwell Tops MLPerf Training 6.0 with... description: NVIDIA achieved a clean sweep in MLPerf Training v6.0, winning every benchmark with its Blackwell platform. The GB300 NVL72 system, featuring 72 Blackwell... canonical: https://daily.dev/posts/nvidia-blackwell-tops-mlperf-training-6-0-with-industry-leading-scale-and-performance-aizek6nop twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: NVIDIA Blackwell Tops MLPerf Training 6.0 with Industry-Leading Scale and Performance | daily.dev og:description: NVIDIA achieved a clean sweep in MLPerf Training v6.0, winning every benchmark with its Blackwell platform. The GB300 NVL72 system, featuring 72 Blackwell... og:url: https://daily.dev/posts/nvidia-blackwell-tops-mlperf-training-6-0-with-industry-leading-scale-and-performance-aizek6nop og:image: https://api.daily.dev/og/posts/Aizek6NOP.png og:image:alt: NVIDIA Blackwell Tops MLPerf Training 6.0 with Industry-Leading Scale and Performance og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # NVIDIA Blackwell Tops MLPerf Training 6.0 with Industry-Leading Scale and Performance **[NVIDIA Developer](https://daily.dev/sources/nvidiadev)** · 11 min read · 0 upvotes · 0 comments ## Summary NVIDIA achieved a clean sweep in MLPerf Training v6.0, winning every benchmark with its Blackwell platform. The GB300 NVL72 system, featuring 72 Blackwell Ultra GPUs connected via NVLink, was the only platform to submit results on all benchmarks including two new ones: DeepSeek-V3 (671B MoE) and GPT-OSS-20B. At peak scale, 8,192 Blackwell GPUs trained DeepSeek-V3 in just 2.02 minutes. Key software innovations driving these results include full-iteration CUDA graphs for token-dropless MoEs (eliminating CPU-GPU sync overhead), CuTe DSL kernel fusions delivering 8%+ end-to-end gains on DeepSeek-V3 and 93% on GPT-OSS, MXFP8 attention blocks, router and hybrid EP optimizations (5x kernel speedup), and 1F1B all-to-all overlap achieving nearly 100% communication overlap. Over three months of continuous optimization, DeepSeek-V3 training throughput improved 1.3x from 1,298 to 1,648 TFLOPS/GPU without hardware changes. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [Fastest, Largest, Strongest: NVIDIA Blackwell Sweeps MLPerf Training 6.0](https://daily.dev/posts/fastest-largest-strongest-nvidia-blackwell-sweeps-mlperf-training-6-0-0jnmk4lcp) · NVIDIA · 0 upvotes · 0 comments - [NVIDIA Extreme Co-Design Delivers New MLPerf Inference Records](https://daily.dev/posts/nvidia-extreme-co-design-delivers-new-mlperf-inference-records-jwjjtv0yx) · NVIDIA Developer · 1 upvotes · 0 comments --- Tags: [#llm](https://daily.dev/tags/llm), [#mixture-of-experts](https://daily.dev/tags/mixture-of-experts) [View this post on daily.dev](https://daily.dev/posts/nvidia-blackwell-tops-mlperf-training-6-0-with-industry-leading-scale-and-performance-aizek6nop) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"NVIDIA Blackwell Tops MLPerf Training 6.0 with Industry-Leading Scale and Performance","url":"https://daily.dev/posts/nvidia-blackwell-tops-mlperf-training-6-0-with-industry-leading-scale-and-performance-aizek6nop","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/nvidia-blackwell-tops-mlperf-training-6-0-with-industry-leading-scale-and-performance-aizek6nop"},"datePublished":"2026-06-16T15:15:20.268Z","dateModified":"2026-06-16T15:16:26.509Z","description":"NVIDIA achieved a clean sweep in MLPerf Training v6.0, winning every benchmark with its Blackwell platform. The GB300 NVL72 system, featuring 72 Blackwell...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/686266eb40119ec046458e48faa3ae53?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/686266eb40119ec046458e48faa3ae53?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"NVIDIA Developer","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"NVIDIA Developer","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/86e45aab42ba48ce83103d01b1119910","url":"https://daily.dev/sources/nvidiadev"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/nvidia-blackwell-tops-mlperf-training-6-0-with-industry-leading-scale-and-performance-aizek6nop","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"llm,mixture-of-experts","timeRequired":"PT11M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"NVIDIA Developer","item":"https://daily.dev/sources/nvidiadev"},{"@type":"ListItem","position":3,"name":"NVIDIA Blackwell Tops MLPerf Training 6.0 with Industry-Leading Scale and Performance"}]} ```