MiniMax M2.5/M2.7 — H200 vs MI325X
Head-to-head AI inference benchmark comparison of H200 (NVIDIA Hopper) and MI325X (AMD CDNA 3) 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.
Setting 34 tok/s/user as the target on MiniMax M2.5/M2.7, H200 produces 3353 tok/s/chip ($0.10 per million tokens) and MI325X produces 2699 ($0.11). H200 is 12% cheaper per token; H200 delivers 24% more tok/s/chip.
At 56 tok/s/user interactivity on MiniMax M2.5/M2.7, H200 delivers 2342 tok/s/chip at $0.14 per million tokens; MI325X delivers 1814 tok/s/chip at $0.17. H200 is 16% cheaper per token; H200 delivers 29% more tok/s/chip at this point.
H200 posts 1681 tok/s/chip for $0.20 per million tokens at 78 tok/s/user on MiniMax M2.5/M2.7; MI325X posts 1009 tok/s/chip for $0.30. H200 is 50% cheaper per token; H200 delivers 67% more tok/s/chip. (Numbers reflect the default 8k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | H200:3352.9MI325X:2698.8 | H200:2341.7MI325X:1814.3 | H200:1680.6MI325X:1009.1 |
| Cost ($/M tok) | H200:$0.101MI325X:$0.113 | H200:$0.145MI325X:$0.168 | H200:$0.202MI325X:$0.303 |
| tok/s/MW | H200:2447382MI325X:1596901 | H200:1709253MI325X:1073566 | H200:1226691MI325X:597118 |
| Concurrency | H200:~44MI325X:~18 | H200:~19MI325X:~8 | H200:~10MI325X:~5 |
Inference Performance
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
Total Tokens per $1 USD (Owning - Hyperscaler) vs. Interactivity
MiniMax M2.5/2.7 230B • FP4 • 8K / 1K • Source: SemiAnalysis InferenceX™
TCO $/chip/hr: H100: 1.17H200: 1.22B200: 1.73B300: 2.26GB200: 1.86GB300: 2.31MI300X: 0.95MI325X: 1.1MI355X: 1.5RTX6000PRO: 0.68
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.
No data available
Please change the model, sequence, precision, date range or chip selection.
Shift+Scroll to zoom • Drag to pan • Double-click to reset • Click a point to pin tooltip