MiniMax M3 428B — B300 vs MI300X Performance per Dollar
Cost per million tokens of B300 (NVIDIA Blackwell) versus MI300X (AMD CDNA 3) on MiniMax M3 428B. 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.
On MiniMax M3 428B at 56 tok/s/user, the per-million math comes out to $0.35 for B300 and $0.93 for MI300X; B300 delivers 165% more output per dollar.
At 99 tok/s/user on MiniMax M3 428B, B300 costs $1.19 per million tokens; MI300X costs $1.87. B300 is 57% more cost-efficient at this operating point.
B300 edges MI300X at 142 tok/s/user on MiniMax M3 428B — $2.60 per million tokens versus $2.87, a 10% cost-per-token gap. (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): B300 $2.34/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 | B300:$0.349MI300X:$0.925 | B300:$1.192MI300X:$1.874 | B300:$2.601MI300X:$2.868 |
| Concurrency | B300:~73MI300X:~27 | B300:~28MI300X:~7 | B300:~7MI300X:~3 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.