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 146 tok/s/user as the target on GLM 5.3 744B, B200 produces 8222 tok/s/chip ($0.06 per million tokens) and GB300 NVL72 produces 13804 ($0.05). GB300 NVL72 is 26% cheaper per token; GB300 NVL72 delivers 68% more tok/s/chip.
At 197 tok/s/user interactivity on GLM 5.3 744B, B200 delivers 6919 tok/s/chip at $0.07 per million tokens; GB300 NVL72 delivers 8969 tok/s/chip at $0.07. B200 is 3% cheaper per token; GB300 NVL72 delivers 30% more tok/s/chip at this point.
B200 posts 4464 tok/s/chip for $0.11 per million tokens at 247 tok/s/user on GLM 5.3 744B; GB300 NVL72 posts 6148 tok/s/chip for $0.10. GB300 NVL72 is 3% cheaper per token; GB300 NVL72 delivers 38% more tok/s/chip. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:8222.1GB300 NVL72:13804.4 | B200:6918.7GB300 NVL72:8969.2 | B200:4464.5GB300 NVL72:6148.2 |
| Cost ($/M tok) | B200:$0.058GB300 NVL72:$0.046 | B200:$0.069GB300 NVL72:$0.072 | B200:$0.108GB300 NVL72:$0.104 |
| tok/s/MW | B200:4808218GB300 NVL72:6511509 | B200:4046020GB300 NVL72:4230769 | B200:2610808GB300 NVL72:2900090 |
| Concurrency | B200:~12GB300 NVL72:~58 | B200:~8GB300 NVL72:~23 | B200:~5GB300 NVL72:~15 |
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.
