Head-to-head AI inference benchmark comparison of H200 (NVIDIA Hopper) and MI355X (AMD CDNA 4) on Muse Spark Open 816B. 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
Throughput at 68 tok/s/user on Muse Spark Open 816B: H200 hits 5674 tok/s/chip, MI355X hits 8974. Per-million costs land at $0.06 and $0.05 respectively. MI355X is 29% cheaper per token; MI355X delivers 58% more tok/s/chip.
H200 / MI355X on Muse Spark Open 816B at 107 tok/s/user: 4227 / 5413 tok/s/chip, $0.08 / $0.08 per million tokens. MI355X is 4% cheaper per token; MI355X delivers 28% more tok/s/chip.
Toward the upper edge of the 30β184 tok/s/user interactivity band, at 145 tok/s/user on Muse Spark Open 816B: H200 runs 2731 tok/s/chip at $0.12/M tokens, MI355X runs 2782 at $0.15/M. H200 is 21% cheaper per token; MI355X delivers 2% more tok/s/chip. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | H200:5673.6MI355X:8974.0 | H200:4227.4MI355X:5412.7 | H200:2731.0MI355X:2781.8 |
| Cost ($/M tok) | H200:$0.060MI355X:$0.046 | H200:$0.080MI355X:$0.077 | H200:$0.124MI355X:$0.150 |
| tok/s/MW | H200:4141321MI355X:4293757 | H200:3085711MI355X:2589817 | H200:1993399MI355X:1330995 |
| Concurrency | H200:~12MI355X:~15 | H200:~7MI355X:~10 | H200:~5MI355X:~1 |
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.
