--- 格式版本: 2 标题: "NVIDIA Blackwell Leads on First Agentic AI Infrastructure Benchmark | NVIDIA Blog" 原文链接: "https://blogs.nvidia.com/blog/nvidia-blackwell-agentperf-artificial-analysis/" 发布日期: "2026-06-30" 发布时间校准状态: "found" 发布时间来源: "llm:strict_original_body" 发布时间证据: "time class=related-news-date nvidia-article-date datetime=2026-06-30T08:00:57-07:00: Jun 30, 2026" 发布时间校准原因: "该日期来自HTML中的time标签，class包含nvidia-article-date，且位于标题附近，符合文章发布时间的特征。" 发布时间校准置信度: "1" 发布时间候选数量: 36 发布时间严格候选数量: 12 发布时间原页读取状态: "原页面来自已抓取 HTML" 发布时间未找到原因: "候选日期无效或 LLM 未确认" 发布时间校准时间: "2026-07-20T11:42:01+08:00" 发现时间: "2026-07-20T09:25:05+08:00" 入库时间: "2026-07-20T04:02:20.855Z" 来源平台: "NVIDIA Blog 搜索" 搜索渠道: "source_template" 搜索词: "https://blogs.nvidia.com/?s=SUE" 匹配关键词: - "SUE" - "NVL72" - "Vera Rubin" 相关厂家: - "NVIDIA" 相关专家: [] 内容类型: "网页" 抓取工具: "AgentKey Scrape" 清洗工具: "AgentKey Markdown + LLM 正文裁剪" 原始附件: [] AI优质: "是" AI打分: 82 AI分档: "高置信优质" AI质检状态: "通过" AI打分理由: "NVIDIA官方发布，直接讨论GB300 NVL72机柜级AI系统架构与Agentic AI基准测试性能，包含72 GPU互联、功耗效率、软件栈优化及量产信号，信息权威且新颖，但正文偏重性能对比，硬件细节略少。" AI质检模型: "qwen3.6-plus" AI质检时间: "2026-07-20T12:02:20+08:00" AI主题相关性: 18 AI来源权威性: 15 AI新颖性: 18 AI技术细节: 16 AI商业部署信号: 10 AI完整性: 5 图片摘要: - "✓ ./assets/img-45465cf2.png | diagram | 展示Agent架构，包含LLM与Harness（Context, Observe, Reason, Act）及外部模块（Prompt, Tools, Memo…" - "★ ./assets/img-2452cff0.jpg | chart | NVIDIA GB300 NVL72与H200在Agentic Coding性能对比，GB300在每兆瓦并发Agent数上领先20倍。" - "✗ ./assets/img-83d2f103.jpg | photo | 其他文章推荐配图" - "✗ ./assets/img-1f6575f1.png | diagram | 其他文章推荐配图" - "✗ ./assets/img-3511966c.jpg | photo | 其他文章推荐配图" - "✗ ./assets/img-e47e99dc.png | other | 其他文章推荐配图" 采集批次: "2026年7月20日9点23分34秒" 采集批次ID: "20260720-092334-062" 去重键: "https://blogs.nvidia.com/blog/nvidia-blackwell-agentperf-artificial-analysis" --- AgentPerf from Artificial Analysis, the industry’s first agentic AI benchmark, gives developers, enterprises and infrastructure providers a clear way to compare systems for agentic AI. In the first round of published results, the [NVIDIA Blackwell Ultra NVL72](https://www.nvidia.com/en-us/data-center/technologies/blackwell-architecture/) platform delivers leading performance across the agentic AI workloads tested, running 20x more agents per megawatt than NVIDIA Hopper. Agentic AI is a fundamentally different workload than conversational AI. A single chat completion is a sprint: one large language model (LLM) call, one response. An agent functions more like a relay: It breaks a goal into many steps and keeps going until the task is done. ![图片](./assets/img-45465cf2.png) Agents chain together multiple LLM calls and tool calls to gather context, observe, reason and act. That results in dozens to hundreds of LLM calls chained together, each passing growing context to the next, with tool calls like code compile and execution, database search and web browsing at every handoff. The complexity isn’t additive; it’s multiplicative. The distinction matters enormously for performance measurement. Existing AI inference benchmarks measure one LLM call: how fast an LLM responds to a single request and how many simultaneous requests a system can handle. They weren’t designed for agentic workloads, where chained LLM calls, tool call delays and growing context stress accelerated computing systems in fundamentally different ways than a single LLM call ever could. For companies building and deploying agents at scale, it’s important to understand how responsive agents are, how many can be deployed simultaneously and how much useful work AI infrastructure can deliver for every dollar and watt invested. ## NVIDIA GB300 NVL72 Runs 20x More Agents per Megawatt In this first round, AgentPerf measures agentic performance with [DeepSeek V4 Pro](https://artificialanalysis.ai/models/deepseek-v4-pro/providers), a large mixture-of-experts (MoE) model that represents the class of frontier models powering today’s most capable agents. On this workload, NVIDIA GB300 NVL72 delivers the highest performance in the benchmark, running up to 20x more agents per megawatt than the NVIDIA HGX H200 system. ![图片](./assets/img-2452cff0.jpg) NVIDIA GB300 NVL72 supports far more concurrent agents per megawatt than NVIDIA H200 at both service-level objectives of 20 and 60 tokens per second per agent. The performance advantage comes from extreme codesign across the full stack. GB300 NVL72 connects 72 GPUs into a single rack-scale system, enabling large MoE models like DeepSeek V4 Pro to distribute model execution efficiently at scale. CUDA kernels accelerate this further by overlapping communication and compute, so the cost of coordinating across experts is absorbed rather than added to latency. NVIDIA TensorRT LLM sustains efficiency as concurrent agent sessions scale. For example, it separates the processing of inputs from the generation of outputs so each can be optimized independently. These results are grounded in a benchmark methodology built from the ground up to reflect how agentic AI actually works in production. ## Artificial Analysis AgentPerf: Built on Real-World Agentic Workloads AgentPerf is built based on real coding agent trajectories: an agent receives a task, reads files, writes and edits code, executes commands and iterates based on the results — all drawn from real public code repositories across 12+ programming languages. The long sequence lengths, tool call patterns and delays are all representative of real-world coding workflows. AgentPerf then measures how many of these agentic tasks a platform can support simultaneously while meeting defined performance thresholds for responsiveness and output token rate. Tool calls are not executed but simulated using representative CPU processing time, so differences in results reflect accelerated computing performance only. The results translate directly into infrastructure decisions: how many concurrent agentic tasks can be run per accelerator and per megawatt of power. For enterprises deploying AI agents at scale, those numbers determine how much productive work a given infrastructure investment can actually deliver. ## NVIDIA Ecosystem Partners Harness Blackwell’s Leading Performance Leading inference providers including Baseten, DeepInfra and Together AI are already serving agentic workloads on frontier models such as [DeepSeek V4 Pro](https://artificialanalysis.ai/models/deepseek-v4-pro/providers) on NVIDIA Blackwell and powering production agentic applications today. [Together AI powers real-time inference for Cursor](https://www.together.ai/blog/learn-how-cursor-partnered-with-together-ai-to-deliver-real-time-low-latency-inference-at-scale), an AI-powered agentic coding platform, on NVIDIA Blackwell. Cursor’s agents debug issues, generate features and execute refactors while developers continue working. DeepInfra powers [Pam.ai](https://pam.ai/), an AI workforce platform for car dealerships, which deploys agents to book service appointments, handle calls and run outbound sales campaigns, entirely on NVIDIA Blackwell. As NVIDIA and the open source ecosystem continue to optimize inference software, performance and efficiency on agentic workloads will only improve. The NVIDIA Vera Rubin architecture is now in full production, bringing the next generation of infrastructure capacity to meet the growing demands of agentic AI at scale. *Dive deeper into AgentPerf’s methodology and NVIDIA’s full-stack optimizations for agentic AI in this [technical blog](https://developer.nvidia.com/blog/nvidia-achieves-leading-agentic-coding-performance-on-first-agentic-ai-benchmark/).* ![图片](./assets/img-45465cf2.png)Agents chain together multiple LLM calls and tool calls to gather context, observe, reason and act. ![图片](./assets/img-2452cff0.jpg)NVIDIA GB300 NVL72 supports far more concurrent agents per megawatt than NVIDIA H200 at both service-level objectives of 20 and 60 tokens per second per agent.