Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and GB300 NVL72 (NVIDIA Blackwell) on GLM 5.3 744B. 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.
AgentX replays real coding-agent sessions rather than fixed-length prompts, so context grows turn over turn and most of each request is served from cache instead of being recomputed. That turns the comparison into a systems question: KV transfer between nodes, prefix-aware routing, and cache capacity all move the curve alongside raw chip throughput. Fixed-sequence workloads stay the clean baseline for kernel and silicon performance, so the two scenarios answer different questions about the same hardware. Learn more about AgentX →
Setting 140 tok/s/user as the target on GLM 5.3 744B, B200 produces 10243 tok/s/chip ($0.05 per million tokens) and GB300 NVL72 produces 16131 ($0.04). GB300 NVL72 is 18% cheaper per token; GB300 NVL72 delivers 57% more tok/s/chip.
At 195 tok/s/user interactivity on GLM 5.3 744B, B200 delivers 7040 tok/s/chip at $0.07 per million tokens; GB300 NVL72 delivers 11269 tok/s/chip at $0.06. GB300 NVL72 is 20% cheaper per token; GB300 NVL72 delivers 60% more tok/s/chip at this point.
B200 posts 4329 tok/s/chip for $0.11 per million tokens at 249 tok/s/user on GLM 5.3 744B; GB300 NVL72 posts 6898 tok/s/chip for $0.09. GB300 NVL72 is 19% cheaper per token; GB300 NVL72 delivers 59% more tok/s/chip. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:10243.3GB300 NVL72:16130.7 | B200:7039.8GB300 NVL72:11268.6 | B200:4328.9GB300 NVL72:6897.6 |
| Cost ($/M tok) | B200:$0.047GB300 NVL72:$0.040 | B200:$0.068GB300 NVL72:$0.057 | B200:$0.111GB300 NVL72:$0.093 |
| tok/s/MW | B200:5990230GB300 NVL72:7608837 | B200:4116854GB300 NVL72:5315391 | B200:2531536GB300 NVL72:3253592 |
| Concurrency | B200:~35GB300 NVL72:~51 | B200:~8GB300 NVL72:~26 | B200:~5GB300 NVL72:~18 |
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
No measurements to plot for this selection. Review the benchmark controls above or adjust quick filters.
