Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) 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
Near the low end of the 84β184 tok/s/user interactivity band, at 109 tok/s/user on Muse Spark Open 816B: B200 runs 21116 tok/s/chip at $0.02/M tokens, MI355X runs 5256 at $0.08/M. B200 is 248% cheaper per token; B200 delivers 302% more tok/s/chip.
Setting 134 tok/s/user as the target on Muse Spark Open 816B, B200 produces 17864 tok/s/chip ($0.03 per million tokens) and MI355X produces 3304 ($0.13). B200 is 369% cheaper per token; B200 delivers 441% more tok/s/chip.
At 159 tok/s/user interactivity on Muse Spark Open 816B, B200 delivers 12700 tok/s/chip at $0.04 per million tokens; MI355X delivers 2510 tok/s/chip at $0.17. B200 is 339% cheaper per token; B200 delivers 406% more tok/s/chip at this point. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:21116.2MI355X:5256.2 | B200:17863.6MI355X:3303.9 | B200:12699.7MI355X:2510.2 |
| Cost ($/M tok) | B200:$0.023MI355X:$0.079 | B200:$0.027MI355X:$0.126 | B200:$0.038MI355X:$0.166 |
| tok/s/MW | B200:12348647MI355X:2514951 | B200:10446557MI355X:1580800 | B200:7426734MI355X:1201046 |
| Concurrency | B200:~36MI355X:~10 | B200:~26MI355X:~2 | B200:~17MI355X:~4 |
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.
