MiniMax M2.5/M2.7 — B300 vs MI325X Performance per Dollar
Cost per million tokens of B300 (NVIDIA Blackwell) versus MI325X (AMD CDNA 3) on MiniMax M2.5/M2.7. 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 edges MI325X at 46 tok/s/user on MiniMax M2.5/M2.7 — $0.13 per million tokens versus $0.22, a 68% cost-per-token gap.
Push MiniMax M2.5/M2.7 to 65 tok/s/user and B300 lands at $0.24 per million tokens against MI325X's $0.48 — B300 pulls ahead by 100%.
B300: $0.47 per million tokens. MI325X: $1.09. Both at 85 tok/s/user on MiniMax M2.5/M2.7, with B300 132% cheaper. (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 · MI325X $1.28/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:$0.129MI325X:$0.217 | B300:$0.240MI325X:$0.479 | B300:$0.472MI325X:$1.094 |
| Concurrency | B300:~455MI325X:~147 | B300:~101MI325X:~38 | B300:~34MI325X:~16 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.