Head-to-head AI inference benchmark comparison of H100 (NVIDIA Hopper) 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
H100 posts 5597 tok/s/chip for $0.06 per million tokens at 97 tok/s/user on MiniMax M3.1 Flash 428B; MI355X posts 49774 tok/s/chip for $0.01. MI355X is 594% cheaper per token; MI355X delivers 789% more tok/s/chip.
Throughput at 161 tok/s/user on MiniMax M3.1 Flash 428B: H100 hits 4304 tok/s/chip, MI355X hits 22528. Per-million costs land at $0.08 and $0.02 respectively. MI355X is 308% cheaper per token; MI355X delivers 423% more tok/s/chip.
H100 / MI355X on MiniMax M3.1 Flash 428B at 224 tok/s/user: 3380 / 17203 tok/s/chip, $0.10 / $0.02 per million tokens. MI355X is 297% cheaper per token; MI355X delivers 409% more tok/s/chip. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | H100:5596.8MI355X:49773.9 | H100:4304.5MI355X:22527.8 | H100:3380.1MI355X:17202.8 |
| Cost ($/M tok) | H100:$0.058MI355X:$0.008 | H100:$0.076MI355X:$0.018 | H100:$0.096MI355X:$0.024 |
| tok/s/MW | H100:4085228MI355X:23815272 | H100:3141967MI355X:10778868 | H100:2467242MI355X:8231027 |
| Concurrency | H100:~8MI355X:~32 | H100:~7MI355X:~13 | H100:~3MI355X:~9 |
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.
