NVIDIA 首个智能体基准，GB300 比 H200 强 23 倍
NVIDIA 发布了首个智能体 AI 基准测试 AgentPerf 的结果。该基准由 Artificial Analysis 推出，测试系统在保持响应性的同时能并发运行多少个智能体。GB300 NVL72 在最低服务层级达到每兆瓦 61.4K 并发智能体，而 H200 仅为 2.6K，性能提升约 23.6 倍。测试模拟了真实编码智能体路径，涵盖 12 种以上编程语言，请求长度从 5K 到 131K token 不等，平均约 27K token。性能提升得益于 72 块 GPU 通过 NVLink 组成机架级系统，以及优化 MoE 专家分配、通信与计算重叠等软件技术。
NVIDIA just posted the first agentic AI benchmark …
NVIDIA just posted the first agentic AI benchmark results where GB300 NVL72 runs up to 20x more coding agents per megawatt than H200.
Older inference benchmarks mostly ask how fast a system can produce tokens after one prompt.
AgentPerf from Artificial Analysis, asks a harder question: how many agents can run at the same time while still feeling responsive.
It tests a harder workload than normal LLM serving because an agent is not one request and one answer, but a long chain of model calls, code edits, command runs, tool delays, and growing context.
The benchmark replays real coding-agent paths from public repos across 12+ programming languages, with request lengths from 5K to 131K tokens and an average near 27K tokens.
NVIDIA says GB300 NVL72 reaches 61.4K concurrent agents per megawatt at the lowest service tier, while H200 reaches 2.6K.
The gain comes from 72 GPUs acting like one rack-scale machine through NVLink, plus software that spreads MoE expert work, overlaps communication with compute, and keeps batches large.
@NVIDIAAIDev