Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and H200 (NVIDIA Hopper) on MiniMax M2.5/M2.7. 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.
B200 posts 9617 tok/s/chip for $0.05 per million tokens at 43 tok/s/user on MiniMax M2.5/M2.7; H200 posts 2816 tok/s/chip for $0.12. B200 is 141% cheaper per token; B200 delivers 241% more tok/s/chip.
Throughput at 72 tok/s/user on MiniMax M2.5/M2.7: B200 hits 3893 tok/s/chip, H200 hits 1855. Per-million costs land at $0.12 and $0.18 respectively. B200 is 48% cheaper per token; B200 delivers 110% more tok/s/chip.
B200 / H200 on MiniMax M2.5/M2.7 at 102 tok/s/user: 2080 / 1011 tok/s/chip, $0.23 / $0.34 per million tokens. B200 is 45% cheaper per token; B200 delivers 106% more tok/s/chip. (Numbers reflect the default 8k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:9616.9H200:2816.4 | B200:3893.2H200:1855.1 | B200:2080.0H200:1011.3 |
| Cost ($/M tok) | B200:$0.050H200:$0.120 | B200:$0.123H200:$0.183 | B200:$0.231H200:$0.335 |
| tok/s/MW | B200:5623907H200:2055797 | B200:2276751H200:1354090 | B200:1216372H200:738156 |
| Concurrency | B200:~582H200:~30 | B200:~32H200:~12 | B200:~15H200:~5 |
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
TCO $/chip/hr: VR200: 3.61H100: 1.17H200: 1.22B200: 1.73B300: 2.26GB200: 1.86GB300: 2.31MI300X: 0.95MI325X: 1.1MI355X: 1.5RTX6000PRO: 0.68Jalapeño (Teacup): 1.47
Source: SemiAnalysis Market July 2026 Pricing Surveys & AI Cloud TCO Model
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate cost per million tokens per decode chip or per prefill chip, rather than per total chip count. This makes direct token cost comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate input throughput per decode chip or per prefill chip, rather than per total chip count. This makes direct input throughput comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate output throughput per decode chip or per prefill chip, rather than per total chip count. This makes direct output throughput comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate power per decode chip or per prefill chip, rather than per total chip count. This makes direct power comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate Joules per decode chip or per prefill chip, rather than per total chip count. This makes direct Joules per token comparison with aggregated configs not an apples-to-apples comparison.
Matching measurements exist, but their chip series are hidden.
Shift+Scroll to zoom • Drag to pan • Double-click to reset • Click a point to pin tooltip
MiniMax M2.5/2.7 230B 8K / 1K Total Tokens per $1 TCO vs. Interactivity