Vera Rubin NVL72 Leads Provisional MLPerf Endpoints Agentic Results
NVIDIA reports that Vera Rubin NVL72 delivered more than 10 times the throughput per provisioned kilowatt of GB300 NVL72 on the Kimi K3 workload in MLPerf Endpoints v1.0, while meeting a shared minimum end-to-end interactivity threshold of 50 tokens per second per user. Its submissions also covered Qwen3.6-35B-A3B and DeepSeek-V4.1-Flash, with checks of generated tool calls, workflow intent and task-solving performance. GB300 results were verified by MLCommons Association, but the Vera Rubin results remain provisional and subject to peer review.
MLPerf Endpoints v1.0 adds agentic coding and enterprise workflows to an API-serving benchmark, so NVIDIA’s results directly address the combined demands of throughput, responsiveness and task correctness for inference infrastructure.
Agentic workloads are demanding more from inference systems. AI agents carry growing conversation histories across turns and generate a mix of reasoning, tool calls, and responses. Inference systems must support users at scale while remaining responsive.
MLPerfEndpoints v1.0 evaluates this balance between serving capacity and responsiveness using coding and enterprise workflows through API endpoints. During performance runs, generated tool calls and workflow intent are checked against reference values. Separate SWE-bench Verified evaluations assess end-to-end task-solving capability. These checks evaluate whether faster serving preserves model quality.
In MLPerfEndpoints v1.0, provisional NVIDIA Vera Rubin NVL72 results show over 10x the throughput per kilowatt of provisioned power of NVIDIA GB300 NVL72 at a shared minimum end-to-end interactivity of 50 tokens per second per user.1
This post presents the benchmark results and outlines the hardware and software optimizations used in the submissions.
1MLPerf® Endpoints v1.0, Kimi K3, Standardized division, Available category, published November 9, 2026. NVIDIA Vera Rubin NVL72, 32 accelerators, result 1.0.1.kimi-k3.agentic_combined.1p, run 9: concurrency = 1,024 queries, TTFT P90 = 4,673.9 ms, system TPS = 41,414.2 tokens/sec, E2E average interactivity = 50.4 tokens/sec per user, operating power = 100.89 kW. NVIDIA GB300 NVL72, 48 accelerators, result 1.0.0.kimi-k3.agentic_combined.1, run 6: concurrency = 48 queries, TTFT P90 = 2,305.1 ms, system TPS = 2,210.4 tokens/sec, E2E average interactivity = 56.2 tokens/sec per user, operating power = 60 kW. Comparison based on throughput per kilowatt at a shared minimum E2E average interactivity of 50 tokens/sec per user. Retrieved from endpoints.mlcommons.org on November 9, 2026. GB300 results verified by MLCommons Association. For the provisional Vera Rubin results: pending MLCommons Association verification, results are subject to change pending peer review. The MLPerf name and logo are registered and unregistered trademarks of MLCommons Association in the United States and other countries. All rights reserved. Unauthorized use is strictly prohibited. See www.mlcommons.org for more information.
#NVIDIA#NVIDIA Vera Rubin NVL72#NVIDIA GB300 NVL72#MLPerf Endpoints v1.0#MLCommons Association#Kimi K3#Qwen3.6-35B-A3B#DeepSeek-V4.1-Flash#SWE-bench Verified