Kimi K2.5/K2.6/K2.7-Code 1T — H200 vs MI325X
Head-to-head AI inference benchmark comparison of H200 (NVIDIA Hopper) and MI325X (AMD CDNA 3) on Kimi K2.5/K2.6/K2.7-Code 1T. 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 24 tok/s/user as the target on Kimi K2.5/K2.6/K2.7-Code 1T, H200 produces 1059 tok/s/chip ($0.32 per million tokens) and MI325X produces 457 ($0.67). H200 is 109% cheaper per token; H200 delivers 131% more tok/s/chip.
At 31 tok/s/user interactivity on Kimi K2.5/K2.6/K2.7-Code 1T, H200 delivers 961 tok/s/chip at $0.35 per million tokens; MI325X delivers 364 tok/s/chip at $0.84. H200 is 138% cheaper per token; H200 delivers 164% more tok/s/chip at this point.
H200 posts 877 tok/s/chip for $0.39 per million tokens at 39 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T; MI325X posts 257 tok/s/chip for $1.19. H200 is 208% cheaper per token; H200 delivers 241% more tok/s/chip. (Numbers reflect the default 8k/1k · int4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | H200:1058.5MI325X:457.3 | H200:961.2MI325X:364.0 | H200:876.8MI325X:257.1 |
| Cost ($/M tok) | H200:$0.320MI325X:$0.668 | H200:$0.353MI325X:$0.839 | H200:$0.387MI325X:$1.189 |
| tok/s/MW | H200:772630MI325X:270608 | H200:701570MI325X:215376 | H200:639998MI325X:152117 |
| Concurrency | H200:~43MI325X:~18 | H200:~28MI325X:~11 | H200:~21MI325X:~6 |
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
Kimi K2.5/2.6/2.7-Code 1T • 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