DeepSeek R1 · Chip comparison

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 51 tok/s/user as the target on DeepSeek R1, H200 produces 1,499 tok/s/chip ($0.23 per million tokens) and MI355X produces 2,191 ($0.19). MI355X is 16% cheaper per token; MI355X delivers 46% more tok/s/chip.

At 90 tok/s/user interactivity on DeepSeek R1, H200 delivers 785 tok/s/chip at $0.43 per million tokens; MI355X delivers 1,391 tok/s/chip at $0.30. MI355X is 31% cheaper per token; MI355X delivers 77% more tok/s/chip at this point.

H200 posts 523 tok/s/chip for $0.65 per million tokens at 129 tok/s/user on DeepSeek R1; MI355X posts 808 tok/s/chip for $0.52. MI355X is 20% cheaper per token; MI355X delivers 55% 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.)

View performance per dollar →

Interpolated from real benchmark data. Edit target interactivity values below to compare at different operating points.

Metric
Interactivity (tok/s/user)
Interactivity (tok/s/user)
Interactivity (tok/s/user)
Throughput (tok/s/chip)
H2001,499MI355X2,191
H200785MI355X1,391
H200523MI355X808
Cost ($/M tok)
H200$0.23MI355X$0.19
H200$0.43MI355X$0.30
H200$0.65MI355X$0.52
tok/s/MW
H2001,094,512MI355X1,048,507
H200573,273MI355X665,678
H200381,794MI355X386,721
Concurrency
H200~54MI355X~34
H200~8MI355X~13
H200~4MI355X~5