MiniMax M2.5/M2.7 — B200 vs MI355X
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and MI355X (AMD CDNA 4) on MiniMax M2.5/M2.7. Latency, throughput, and cost across LLM workloads. Use the chart controls below to switch sequences, precisions, and metrics — same interactions as the main inference chart.
Near the low end of the 19–117 tok/s/user interactivity band, at 43 tok/s/user on MiniMax M2.5/M2.7: B200 runs 5595 tok/s/GPU at $0.10/M tokens, MI355X runs 3663 at $0.11/M. B200 is 16% cheaper per token; B200 delivers 53% more tok/s/GPU.
Setting 68 tok/s/user as the target on MiniMax M2.5/M2.7, B200 produces 2455 tok/s/GPU ($0.22 per million tokens) and MI355X produces 1151 ($0.36). B200 is 63% cheaper per token; B200 delivers 113% more tok/s/GPU.
At 93 tok/s/user interactivity on MiniMax M2.5/M2.7, B200 delivers 1103 tok/s/GPU at $0.47 per million tokens; MI355X delivers 548 tok/s/GPU at $0.73. B200 is 56% cheaper per token; B200 delivers 101% more tok/s/GPU at this point. (Numbers reflect the default 1k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | B200:5594.6MI355X:3663.4 | B200:2455.2MI355X:1150.9 | B200:1102.9MI355X:548.5 |
| Cost ($/M tok) | B200:$0.097MI355X:$0.112 | B200:$0.221MI355X:$0.360 | B200:$0.469MI355X:$0.731 |
| tok/s/MW | B200:2578149MI355X:1382396 | B200:1131420MI355X:434286 | B200:508248MI355X:206964 |
| Concurrency | B200:~272MI355X:~256 | B200:~75MI355X:~35 | B200:~24MI355X:~8 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.