Kimi K2.5/K2.6/K2.7-Code 1T — H200 vs MI355X
Head-to-head AI inference benchmark comparison of H200 (NVIDIA Hopper) and MI355X (AMD CDNA 4) 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.
Throughput at 32 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T: H200 hits 949 tok/s/chip, MI355X hits 1594. Per-million costs land at $0.36 and $0.26 respectively. MI355X is 37% cheaper per token; MI355X delivers 68% more tok/s/chip.
H200 / MI355X on Kimi K2.5/K2.6/K2.7-Code 1T at 48 tok/s/user: 792 / 1122 tok/s/chip, $0.43 / $0.37 per million tokens. MI355X is 15% cheaper per token; MI355X delivers 42% more tok/s/chip.
Toward the upper edge of the 17–78 tok/s/user interactivity band, at 63 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T: H200 runs 655 tok/s/chip at $0.52/M tokens, MI355X runs 840 at $0.50/M. MI355X is 4% cheaper per token; MI355X delivers 28% more tok/s/chip. (Numbers reflect the default 8k/1k · int4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | H200:949.3MI355X:1594.5 | H200:792.1MI355X:1121.6 | H200:655.2MI355X:840.3 |
| Cost ($/M tok) | H200:$0.357MI355X:$0.261 | H200:$0.428MI355X:$0.371 | H200:$0.517MI355X:$0.496 |
| tok/s/MW | H200:692906MI355X:762896 | H200:578178MI355X:536643 | H200:478223MI355X:402067 |
| Concurrency | H200:~27MI355X:~23 | H200:~15MI355X:~11 | H200:~10MI355X:~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