Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and MI355X (AMD CDNA 4) on DeepSeek V4.1 Flash 552B. Latency, throughput, and cost across LLM workloads. Use the chart controls below to switch sequences, precisions, and metrics — 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 →
B200 posts 87242 tok/s/chip for $0.01 per million tokens at 19 tok/s/user on DeepSeek V4.1 Flash 552B; MI355X posts 9485 tok/s/chip for $0.04. B200 is 697% cheaper per token; B200 delivers 820% more tok/s/chip.
Throughput at 24 tok/s/user on DeepSeek V4.1 Flash 552B: B200 hits 87242 tok/s/chip, MI355X hits 6984. Per-million costs land at $0.01 and $0.06 respectively. B200 is 983% cheaper per token; B200 delivers 1149% more tok/s/chip.
B200 / MI355X on DeepSeek V4.1 Flash 552B at 30 tok/s/user: 87242 / 3578 tok/s/chip, $0.01 / $0.12 per million tokens. B200 is 2014% cheaper per token; B200 delivers 2339% more tok/s/chip. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:87242.3MI355X:9485.1 | B200:87242.3MI355X:6983.9 | B200:87242.3MI355X:3577.6 |
| Cost ($/M tok) | B200:$0.006MI355X:$0.044 | B200:$0.006MI355X:$0.060 | B200:$0.006MI355X:$0.116 |
| tok/s/MW | B200:51018864MI355X:4538332 | B200:51018864MI355X:3341561 | B200:51018864MI355X:1711792 |
| Concurrency | B200:~64MI355X:~17 | B200:~64MI355X:~10 | B200:~64MI355X:~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.
