MiniMax M3 428B — B300 vs MI355X Performance per Dollar
Cost per million tokens of B300 (NVIDIA Blackwell) versus MI355X (AMD CDNA 4) 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.
B300: $0.44 per million tokens. MI355X: $0.28. Both at 95 tok/s/user on MiniMax M3 428B, with MI355X 60% cheaper.
Around the middle of the 12–348 tok/s/user interactivity band — at 180 tok/s/user — B300 runs $0.85 per million tokens on MiniMax M3 428B while MI355X runs $0.72. MI355X is the cheaper choice by 18%.
On MiniMax M3 428B at 264 tok/s/user, the per-million math comes out to $1.70 for B300 and $1.97 for MI355X; B300 delivers 16% 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): B300 $2.26/chip/hr · MI355X $1.50/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 | B300:$0.443MI355X:$0.277 | B300:$0.852MI355X:$0.722 | B300:$1.698MI355X:$1.972 |
| Concurrency | B300:~45MI355X:~35 | B300:~8MI355X:~7 | B300:~6MI355X:~2 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.