GLM 5/5.1 — B300 vs GB300 NVL72
Head-to-head AI inference benchmark comparison of B300 (NVIDIA Blackwell) and GB300 NVL72 (NVIDIA Blackwell) 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.
B300 / GB300 NVL72 on GLM 5/5.1 at 31 tok/s/user: 1037 / 3824 tok/s/GPU, $0.63 / $0.19 per million tokens. GB300 NVL72 is 226% cheaper per token; GB300 NVL72 delivers 269% more tok/s/GPU.
Around the middle of the 19–67 tok/s/user interactivity band, at 43 tok/s/user on GLM 5/5.1: B300 runs 790 tok/s/GPU at $0.83/M tokens, GB300 NVL72 runs 276 at $2.67/M. B300 is 223% cheaper per token; B300 delivers 187% more tok/s/GPU.
Setting 55 tok/s/user as the target on GLM 5/5.1, B300 produces 628 tok/s/GPU ($1.02 per million tokens) and GB300 NVL72 produces 93 ($7.90). B300 is 673% cheaper per token; B300 delivers 574% more tok/s/GPU. (Numbers reflect the default 1k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | B300:1037.0GB300 NVL72:3824.1 | B300:790.0GB300 NVL72:275.7 | B300:628.1GB300 NVL72:93.2 |
| Cost ($/M tok) | B300:$0.627GB300 NVL72:$0.193 | B300:$0.826GB300 NVL72:$2.674 | B300:$1.023GB300 NVL72:$7.904 |
| tok/s/MW | B300:477859GB300 NVL72:1820984 | B300:364046GB300 NVL72:131293 | B300:289444GB300 NVL72:44379 |
| Concurrency | B300:~140GB300 NVL72:~6137 | B300:~76GB300 NVL72:~258 | B300:~46GB300 NVL72:~67 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.