Llama 3.3 70B — H200 vs MI300X Performance per Dollar
Cost per million tokens of H200 (NVIDIA Hopper) versus MI300X (AMD CDNA 3) 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.
H200: $0.09 per million tokens. MI300X: $0.19. Both at 44 tok/s/user on Llama 3.3 70B, with H200 106% cheaper.
Around the middle of the 22–112 tok/s/user interactivity band — at 67 tok/s/user — H200 runs $0.14 per million tokens on Llama 3.3 70B while MI300X runs $0.32. H200 is the cheaper choice by 127%.
On Llama 3.3 70B at 90 tok/s/user, the per-million math comes out to $0.22 for H200 and $0.63 for MI300X; H200 delivers 185% more output per dollar. (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.)
Chip pricing (owning hyperscaler): H200 $1.22/chip/hr · MI300X $0.95/chip/hr. Source: SemiAnalysis Market July 2026 Pricing Surveys & AI Cloud TCO Model.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Dollar per Million Tokens | H200:$0.091MI300X:$0.188 | H200:$0.141MI300X:$0.320 | H200:$0.222MI300X:$0.632 |
| Concurrency | H200:~96MI300X:~64 | H200:~64MI300X:~32 | H200:~37MI300X:~20 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.