DeepSeek R1 — H200 vs MI355X
Head-to-head AI inference benchmark comparison of H200 (NVIDIA Hopper) and MI355X (AMD CDNA 4) on DeepSeek R1. 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 50 tok/s/user as the target on DeepSeek R1, H200 produces 1591 tok/s/chip ($0.21 per million tokens) and MI355X produces 2109 ($0.20). MI355X is 8% cheaper per token; MI355X delivers 33% more tok/s/chip.
At 89 tok/s/user interactivity on DeepSeek R1, H200 delivers 569 tok/s/chip at $0.60 per million tokens; MI355X delivers 1185 tok/s/chip at $0.35. MI355X is 69% cheaper per token; MI355X delivers 108% more tok/s/chip at this point.
H200 posts 292 tok/s/chip for $1.16 per million tokens at 128 tok/s/user on DeepSeek R1; MI355X posts 908 tok/s/chip for $0.46. MI355X is 153% cheaper per token; MI355X delivers 211% more tok/s/chip. (Numbers reflect the default 8k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | H200:1590.9MI355X:2109.2 | H200:569.0MI355X:1184.8 | H200:292.1MI355X:907.7 |
| Cost ($/M tok) | H200:$0.213MI355X:$0.198 | H200:$0.596MI355X:$0.352 | H200:$1.160MI355X:$0.459 |
| tok/s/MW | H200:1161268MI355X:1009196 | H200:415295MI355X:566886 | H200:213203MI355X:434290 |
| Concurrency | H200:~73MI355X:~34 | H200:~6MI355X:~217 | H200:~21MI355X:~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
DeepSeek R1 0528 671B • 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