NVIDIA Vera Rubin NVL72 智能体推理首测：每兆瓦吞吐最高约 5.7 倍于 GB300
Initial results for NVIDIA's Vera Rubin NVL72 are now live on AA-AgentPerf, with up to ~5.7x the total throughput per megawatt of its predecessor GB300 NVL72 on
Artificial Analysis 在 AA-AgentPerf 上发布 NVIDIA Vera Rubin NVL72 的初步结果，在 DeepSeek V4 Pro 上每兆瓦总吞吐最高约为前代 GB300 NVL72 的 5.7 倍，DeepSeek V4.1 Flash 约 5.4 倍，GLM-5.2 约 3.0 倍，Kimi K3 约 2.8 倍。Vera Rubin 是 NVIDIA Blackwell 的继任者，NVL72 为机架级系统，含 72 颗 Rubin GPU 和 36 颗 Vera CPU，每颗 GPU 内存与 Blackwell Ultra 同为 288 GB，但内存带宽为 2.4 倍、NVLink 带宽 1.7 倍、密集 NVFP4 算力 2.3 倍。AA-AgentPerf 通过回放真实智能体编码会话衡量推理性能，Vera Rubin 在 DeepSeek V4.1 Flash 上创下 906 tok/s 的单用户速度纪录。
推荐理由
AI 评分 65
AA-AgentPerf 首次给出 Vera Rubin NVL72 的智能体推理实测，读者可据此比较两代机架系统的吞吐与单用户速度取舍。
主题
正文
当前显示原文，中文全文译文尚未就绪。
Initial results for NVIDIA's Vera Rubin NVL72 are now live on AA-AgentPerf, with up to ~5.7x the total throughput per megawatt of its predecessor GB300 NVL72 on DeepSeek V4 Pro, ~5.4x on DeepSeek V4.1 Flash, ~3.0x on GLM-5.2 and ~2.8x on Kimi K3
Vera Rubin is NVIDIA's successor to Blackwell, and Vera Rubin NVL72 is its rack-scale system, with 72 Rubin GPUs and 36 Vera CPUs. Each Rubin GPU has the same 288 GB of memory as Blackwell Ultra, but 2.4x the memory bandwidth, 1.7x the NVLink bandwidth and 2.3x the dense NVFP4 compute. NVIDIA submitted every Vera Rubin result, running TensorRT-LLM and vLLM with NVIDIA Dynamo
AA-AgentPerf is our agentic inference performance benchmark. It replays real agentic coding sessions, including long cached contexts and tool-call delays, to give a realistic measure of inference performance in day-to-day agentic workloads. We measure total system throughput per megawatt of measured GPU power, and per-agent output speed (p25 across requests) at varying concurrency levels, to show the trade-off between system utilization and individual user experience
Initial results:
➤ Vera Rubin's lead is largest at faster per-user speeds, where GB300's throughput falls away. On DeepSeek V4 Pro, going from ~120 to ~165 tok/s per user costs GB300 81% of its throughput per MW, compared to 16% for Vera Rubin
➤ At peak throughput the two are closer, with Vera Rubin at 272M tokens/s per MW on DeepSeek V4 Pro vs 239M for GB300 (1.14x), 571M vs 395M on DeepSeek V4.1 Flash (1.45x), 272M vs 201M on GLM-5.2 (1.35x) and 70M vs 56M on Kimi K3 (1.26x). We expect the gap between the systems to widen as Vera Rubin kernels and other serving software are optimized
➤ Vera Rubin sets the fastest per-user speeds on our leaderboard for DeepSeek V4.1 Flash (906 tok/s, the fastest result for any model), GLM-5.2 (408 tok/s) and Kimi K3 (255 tok/s)
Congratulations @nvidia on the strong results!