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 β 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 β
At 148 tok/s/user interactivity on GLM 5.3 744B, B300 delivers 9932 tok/s/chip at $0.06 per million tokens; GB300 NVL72 delivers 16193 tok/s/chip at $0.04. GB300 NVL72 is 60% cheaper per token; GB300 NVL72 delivers 63% more tok/s/chip at this point.
B300 posts 7262 tok/s/chip for $0.09 per million tokens at 205 tok/s/user on GLM 5.3 744B; GB300 NVL72 posts 10438 tok/s/chip for $0.06. GB300 NVL72 is 41% cheaper per token; GB300 NVL72 delivers 44% more tok/s/chip.
Throughput at 262 tok/s/user on GLM 5.3 744B: B300 hits 4259 tok/s/chip, GB300 NVL72 hits 4232. Per-million costs land at $0.15 and $0.15 respectively. B300 is 3% cheaper per token; throughput per chip is essentially tied. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B300:9932.1GB300 NVL72:16193.4 | B300:7262.3GB300 NVL72:10437.6 | B300:4259.3GB300 NVL72:4231.7 |
| Cost ($/M tok) | B300:$0.063GB300 NVL72:$0.040 | B300:$0.086GB300 NVL72:$0.061 | B300:$0.147GB300 NVL72:$0.152 |
| tok/s/MW | B300:5227396GB300 NVL72:7638404 | B300:3822269GB300 NVL72:4923420 | B300:2241722GB300 NVL72:1996075 |
| Concurrency | B300:~14GB300 NVL72:~55 | B300:~8GB300 NVL72:~20 | B300:~4GB300 NVL72:~4 |
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.
