Head-to-head AI inference benchmark comparison of GB200 NVL72 (NVIDIA Blackwell) and MI355X (AMD CDNA 4) on MiniMax M3.1 Flash 428B. Latency, throughput, and cost across LLM workloads. Use the chart controls below to switch sequences, precisions, and metrics β same interactions as the main inference chart.
AgentX replays real coding-agent sessions rather than fixed-length prompts, so context grows turn over turn and most of each request is served from cache instead of being recomputed. That turns the comparison into a systems question: KV transfer between nodes, prefix-aware routing, and cache capacity all move the curve alongside raw chip throughput. Fixed-sequence workloads stay the clean baseline for kernel and silicon performance, so the two scenarios answer different questions about the same hardware. Learn more about AgentX β
At 157 tok/s/user interactivity on MiniMax M3.1 Flash 428B, GB200 NVL72 delivers 42232 tok/s/chip at $0.01 per million tokens; MI355X delivers 23069 tok/s/chip at $0.02. GB200 NVL72 is 48% cheaper per token; GB200 NVL72 delivers 83% more tok/s/chip at this point.
GB200 NVL72 posts 20538 tok/s/chip for $0.03 per million tokens at 281 tok/s/user on MiniMax M3.1 Flash 428B; MI355X posts 13537 tok/s/chip for $0.03. GB200 NVL72 is 22% cheaper per token; GB200 NVL72 delivers 52% more tok/s/chip.
Throughput at 405 tok/s/user on MiniMax M3.1 Flash 428B: GB200 NVL72 hits 11295 tok/s/chip, MI355X hits 7542. Per-million costs land at $0.05 and $0.06 respectively. GB200 NVL72 is 21% cheaper per token; GB200 NVL72 delivers 50% more tok/s/chip. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | GB200 NVL72:42231.9MI355X:23069.2 | GB200 NVL72:20538.5MI355X:13537.3 | GB200 NVL72:11294.8MI355X:7542.1 |
| Cost ($/M tok) | GB200 NVL72:$0.012MI355X:$0.018 | GB200 NVL72:$0.025MI355X:$0.031 | GB200 NVL72:$0.046MI355X:$0.055 |
| tok/s/MW | GB200 NVL72:22583926MI355X:11037876 | GB200 NVL72:10983141MI355X:6477176 | GB200 NVL72:6039981MI355X:3608657 |
| Concurrency | GB200 NVL72:~25MI355X:~14 | GB200 NVL72:~11MI355X:~7 | GB200 NVL72:~6MI355X:~3 |
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
No measurements to plot for this selection. Review the benchmark controls above or adjust quick filters.
