Kimi K2.5/K2.6/K2.7-Code 1T — GB200 NVL72 vs MI355X
Head-to-head AI inference benchmark comparison of GB200 NVL72 (NVIDIA Blackwell) 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.
Near the low end of the 26–123 tok/s/user interactivity band, at 50 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T: GB200 NVL72 runs 13143 tok/s/chip at $0.04/M tokens, MI355X runs 3146 at $0.13/M. GB200 NVL72 is 237% cheaper per token; GB200 NVL72 delivers 318% more tok/s/chip.
Setting 74 tok/s/user as the target on Kimi K2.5/K2.6/K2.7-Code 1T, GB200 NVL72 produces 6866 tok/s/chip ($0.08 per million tokens) and MI355X produces 2269 ($0.18). GB200 NVL72 is 144% cheaper per token; GB200 NVL72 delivers 203% more tok/s/chip.
At 99 tok/s/user interactivity on Kimi K2.5/K2.6/K2.7-Code 1T, GB200 NVL72 delivers 1548 tok/s/chip at $0.33 per million tokens; MI355X delivers 1618 tok/s/chip at $0.26. MI355X is 30% cheaper per token; MI355X delivers 4% more tok/s/chip at this point. (Numbers reflect the default 8k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | GB200 NVL72:13143.1MI355X:3145.5 | GB200 NVL72:6866.4MI355X:2269.2 | GB200 NVL72:1548.2MI355X:1617.8 |
| Cost ($/M tok) | GB200 NVL72:$0.039MI355X:$0.132 | GB200 NVL72:$0.075MI355X:$0.184 | GB200 NVL72:$0.334MI355X:$0.258 |
| tok/s/MW | GB200 NVL72:7028421MI355X:1505042 | GB200 NVL72:3671864MI355X:1085724 | GB200 NVL72:827888MI355X:774072 |
| Concurrency | GB200 NVL72:~1229MI355X:~29 | GB200 NVL72:~697MI355X:~14 | GB200 NVL72:~256MI355X:~8 |
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