Head-to-head AI inference benchmark comparison of B300 (NVIDIA Blackwell) and H200 (NVIDIA Hopper) on MiniMax M3.1 Flash 428B. 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
Near the low end of the 9β294 tok/s/user interactivity band, at 80 tok/s/user on MiniMax M3.1 Flash 428B: B300 runs 62894 tok/s/chip at $0.01/M tokens, H200 runs 9878 at $0.03/M. B300 is 244% cheaper per token; B300 delivers 537% more tok/s/chip.
Setting 151 tok/s/user as the target on MiniMax M3.1 Flash 428B, B300 produces 60991 tok/s/chip ($0.01 per million tokens) and H200 produces 7177 ($0.05). B300 is 359% cheaper per token; B300 delivers 750% more tok/s/chip.
At 223 tok/s/user interactivity on MiniMax M3.1 Flash 428B, B300 delivers 37840 tok/s/chip at $0.02 per million tokens; H200 delivers 4231 tok/s/chip at $0.08. B300 is 383% cheaper per token; B300 delivers 794% more tok/s/chip at this point. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B300:62894.4H200:9878.0 | B300:60990.9H200:7177.4 | B300:37839.9H200:4231.4 |
| Cost ($/M tok) | B300:$0.010H200:$0.034 | B300:$0.010H200:$0.047 | B300:$0.017H200:$0.080 |
| tok/s/MW | B300:33102319H200:7210255 | B300:32100496H200:5238983 | B300:19915762H200:3088604 |
| Concurrency | B300:~54H200:~16 | B300:~39H200:~10 | B300:~20H200:~5 |
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.
