Head-to-head AI inference benchmark comparison of H200 (NVIDIA Hopper) and MI355X (AMD CDNA 4) on Kimi K3 2.8T. 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
Near the low end of the 2β155 tok/s/user interactivity band, at 40 tok/s/user on Kimi K3 2.8T: H200 runs 70 tok/s/chip at $4.85/M tokens, MI355X runs 10268 at $0.04/M. MI355X is 11853% cheaper per token; MI355X delivers 14596% more tok/s/chip.
Setting 78 tok/s/user as the target on Kimi K3 2.8T, H200 produces 70 tok/s/chip ($4.85 per million tokens) and MI355X produces 4327 ($0.10). MI355X is 4937% cheaper per token; MI355X delivers 6093% more tok/s/chip.
At 117 tok/s/user interactivity on Kimi K3 2.8T, H200 delivers 70 tok/s/chip at $4.85 per million tokens; MI355X delivers 3247 tok/s/chip at $0.13. MI355X is 3680% cheaper per token; MI355X delivers 4548% more tok/s/chip at this point. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | H200:69.9MI355X:10268.2 | H200:69.9MI355X:4327.0 | H200:69.9MI355X:3247.3 |
| Cost ($/M tok) | H200:$4.850MI355X:$0.041 | H200:$4.850MI355X:$0.096 | H200:$4.850MI355X:$0.128 |
| tok/s/MW | H200:51001MI355X:4913013 | H200:51001MI355X:2070354 | H200:51001MI355X:1553749 |
| Concurrency | H200:~1MI355X:~22 | H200:~1MI355X:~7 | H200:~1MI355X:~5 |
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.
