Head-to-head AI inference benchmark comparison of B300 (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 →
At 168 tok/s/user interactivity on GLM 5.3 744B, B300 delivers 7521 tok/s/chip at $0.08 per million tokens; GB300 NVL72 delivers 10370 tok/s/chip at $0.06. GB300 NVL72 is 35% cheaper per token; GB300 NVL72 delivers 38% more tok/s/chip at this point.
B300 posts 5343 tok/s/chip for $0.12 per million tokens at 211 tok/s/user on GLM 5.3 744B; GB300 NVL72 posts 8824 tok/s/chip for $0.07. GB300 NVL72 is 62% cheaper per token; GB300 NVL72 delivers 65% more tok/s/chip.
Throughput at 255 tok/s/user on GLM 5.3 744B: B300 hits 4219 tok/s/chip, GB300 NVL72 hits 5353. Per-million costs land at $0.15 and $0.12 respectively. GB300 NVL72 is 24% cheaper per token; GB300 NVL72 delivers 27% more tok/s/chip. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B300:7521.3GB300 NVL72:10369.6 | B300:5342.9GB300 NVL72:8823.9 | B300:4219.0GB300 NVL72:5353.1 |
| Cost ($/M tok) | B300:$0.083GB300 NVL72:$0.062 | B300:$0.117GB300 NVL72:$0.073 | B300:$0.149GB300 NVL72:$0.120 |
| tok/s/MW | B300:3958572GB300 NVL72:4891303 | B300:2812059GB300 NVL72:4162226 | B300:2220540GB300 NVL72:2525024 |
| Concurrency | B300:~10GB300 NVL72:~49 | B300:~6GB300 NVL72:~30 | B300:~4GB300 NVL72:~10 |
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.
