gpt-oss 120B — H100 vs MI300X Performance per Dollar
Cost per million tokens of H100 (NVIDIA Hopper) versus MI300X (AMD CDNA 3) on gpt-oss 120B. 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 112 tok/s/user on gpt-oss 120B, H100 costs $0.13 per million tokens; MI300X costs $0.20. H100 is 50% more cost-efficient at this operating point.
H100 edges MI300X at 158 tok/s/user on gpt-oss 120B — $0.24 per million tokens versus $0.37, a 57% cost-per-token gap.
Push gpt-oss 120B to 203 tok/s/user and H100 lands at $0.41 per million tokens against MI300X's $0.97 — H100 pulls ahead by 138%. (Numbers reflect the default 1k/1k · fp4 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 · MI300X $1.12/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:$0.131MI300X:$0.195 | H100:$0.238MI300X:$0.373 | H100:$0.409MI300X:$0.975 |
| Concurrency | H100:~64MI300X:~15 | H100:~28MI300X:~5 | H100:~10MI300X:~7 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.