DeepSeek R1 · Performance per Dollar

DeepSeek R1 — H100 vs MI325X Performance per Dollar

Cost per million tokens of H100 (NVIDIA Hopper) versus MI325X (AMD CDNA 3) on DeepSeek R1. Owning-hyperscaler TCO normalized by output tokens — performance per dollar across LLM workloads. Pick the more cost-efficient SKU at every target interactivity level. Use the chart controls below to switch sequences, precisions, and metrics — same interactions as the main inference chart.

At 39 tok/s/user on DeepSeek R1, H100 costs $1.33 per million tokens; MI325X costs $1.03. MI325X is 29% more cost-efficient at this operating point.

MI325X edges H100 at 50 tok/s/user on DeepSeek R1 — $1.57 per million tokens versus $2.20, a 40% cost-per-token gap.

Push DeepSeek R1 to 60 tok/s/user and H100 lands at $2.74 per million tokens against MI325X's $2.67 — MI325X pulls ahead by 3%. (Numbers reflect the default 1k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)

GPU pricing (owning hyperscaler): H100 $1.30/GPU/hr · MI325X $1.28/GPU/hr. Source: SemiAnalysis Market August 2025 Pricing Surveys & AI Cloud TCO Model.

View full latency + throughput comparison →

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)
Dollar per Million Tokens
H100:$1.328MI325X:$1.027
H100:$2.197MI325X:$1.566
H100:$2.739MI325X:$2.672
Concurrency
H100:~624MI325X:~39
H100:~203MI325X:~19
H100:~128MI325X:~9

Inference Performance

Inference performance metrics across different models, hardware configurations, and serving parameters.