Llama 3.3 70B · Performance per Dollar

Llama 3.3 70B — H200 vs MI355X Performance per Dollar

Cost per million tokens of H200 (NVIDIA Hopper) versus MI355X (AMD CDNA 4) on Llama 3.3 70B. Large-hyperscaler-volume ownership TCO normalized by total 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.

On Llama 3.3 70B at 37 tok/s/user, the per-million math comes out to $0.10 for H200 and $0.096 for MI355X; MI355X delivers 5% more total tokens per dollar.

At 61 tok/s/user on Llama 3.3 70B, H200 costs $0.14 per million tokens; MI355X costs $0.16. H200 delivers 14% more tokens per dollar at this operating point.

H200 edges MI355X at 85 tok/s/user on Llama 3.3 70B — $0.21 per million tokens versus $0.39, a 48% lower cost per token. (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.)

Chip pricing (owning hyperscaler): H200 $1.22/chip/hr · MI355X $1.50/chip/hr. Source: SemiAnalysis Market July 2026 Pricing Surveys & AI Cloud TCO Model.

View full latency and 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)
Dollars per Million Tokens
H200$0.10MI355X$0.096
H200$0.14MI355X$0.16
H200$0.21MI355X$0.39
Concurrency
H200~32MI355X~18
H200~20MI355X~22
H200~8MI355X~6
Llama 3.3 70B: H200 versus MI355X cost per million tokens at matched interactivity levels
H200 versus MI355X cost per million tokens for this comparison's canonical default workload. Lower cost indicates better performance per dollar.
Open full-size image ↗