GLM 5/5.1 — GB300 NVL72 vs H200
Head-to-head AI inference benchmark comparison of GB300 NVL72 (NVIDIA Blackwell) and H200 (NVIDIA Hopper) on GLM 5/5.1. Latency, throughput, and cost across LLM workloads. Use the chart controls below to switch sequences, precisions, and metrics — same interactions as the main inference chart.
GB300 NVL72 hits 6679 tok/s/GPU for $0.11 per million tokens at 69 tok/s/user on GLM 5/5.1. No H200 data at this operating point.
GB300 NVL72: 811 tok/s/GPU, $0.91 per million tokens at 121 tok/s/user on GLM 5/5.1. H200 is unmeasured here.
At 173 tok/s/user on GLM 5/5.1, GB300 NVL72 delivers 151 tok/s/GPU at $4.87 per million tokens; H200 hasn't been benchmarked at this target. (Numbers reflect the default 1k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | GB300 NVL72:6679.2H200:— | GB300 NVL72:810.6H200:— | GB300 NVL72:151.3H200:— |
| Cost ($/M tok) | GB300 NVL72:$0.110H200:— | GB300 NVL72:$0.910H200:— | GB300 NVL72:$4.870H200:— |
| tok/s/MW | GB300 NVL72:3180562H200:— | GB300 NVL72:385983H200:— | GB300 NVL72:72026H200:— |
| Concurrency | GB300 NVL72:~1221H200:— | GB300 NVL72:~166H200:— | GB300 NVL72:~23H200:— |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.