PulseAugur
实时 18:06:29
English(EN) NVIDIA Vera Rubin NVL72 Delivers Leading Performance in MLPerf Inference v6.1 Debut

NVIDIA Vera Rubin NVL72 系统在 MLPerf Inference v6.1 性能测试中首次亮相并取得领先

NVIDIA 宣布其新的 Vera Rubin NVL72 系统在 MLPerf Inference v6.1 基准测试中取得了领先性能。该系统在 Qwen3-VL 和 DeepSeek-R1 等要求严苛的模型上,吞吐量比前代 GB300 NVL72 高出 3.7 倍。这一性能归功于跨硬件和软件的全栈协同设计,包括增强的 Tensor Cores、Transformer Engine 和优化的互连。NVIDIA 还强调了该系统高效的扩展能力,在多机架提交中实现了 99% 的效率,并有望通过生成更多 token 和以更低成本服务更多用户来显著改善 AI 推理的经济性。 AI

影响 为 AI 推理硬件设定了新的性能基准,有望降低 AI 部署的成本并提高效率。

排序理由 新硬件系统首次提交至公认的行业基准测试。 [lever_c_demoted from research: ic=1 ai=0.7]

在 NVIDIA Blog 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

NVIDIA Vera Rubin NVL72 系统在 MLPerf Inference v6.1 性能测试中首次亮相并取得领先

本文如何被排名

Signal score
14 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
新硬件系统首次提交至公认的行业基准测试。 [lever_c_demoted from research: ic=1 ai=0.7]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
infra, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准

报道来源 [1]

  1. NVIDIA Blog TIER_1 English(EN) · Zhihan Jiang ·

    NVIDIA Vera Rubin NVL72 在 MLPerf Inference v6.1 首次亮相中展现领先性能

    System performance, efficient infrastructure scaling and continuous software optimization are key levers that determine AI inference economics. Higher system performance means more tokens generated, resulting in higher revenue. Efficient scaling means throughput grows proportiona…