Llama 3.3 70B — H100 vs MI355X Performance per Dollar
Cost per million tokens of H100 (NVIDIA Hopper) versus MI355X (AMD CDNA 4) on Llama 3.3 70B. 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 53 tok/s/user on Llama 3.3 70B, H100 costs $0.25 per million tokens; MI355X costs $0.18. MI355X is 39% more cost-efficient at this operating point.
MI355X edges H100 at 72 tok/s/user on Llama 3.3 70B — $0.25 per million tokens versus $0.45, a 84% cost-per-token gap.
Push Llama 3.3 70B to 91 tok/s/user and H100 lands at $1.19 per million tokens against MI355X's $0.42 — MI355X pulls ahead by 182%. (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 · MI355X $1.48/GPU/hr. Source: SemiAnalysis Market August 2025 Pricing Surveys & AI Cloud TCO Model.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Dollar per Million Tokens | H100:$0.249MI355X:$0.179 | H100:$0.454MI355X:$0.247 | H100:$1.194MI355X:$0.424 |
| Concurrency | H100:~64MI355X:~47 | H100:~50MI355X:~53 | H100:~14MI355X:~22 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.