Kimi K3.1 Pro — GB200 NVL72 vs MI355X
Head-to-head AI inference benchmark comparison of GB200 NVL72 (NVIDIA Blackwell) and MI355X (AMD CDNA 4) on Kimi K3.1 Pro. 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 →
Near the low end of the 27–175 tok/s/user interactivity band, at 63 tok/s/user on Kimi K3.1 Pro: GB200 NVL72 runs 8352 tok/s/chip at $0.06/M tokens, MI355X runs 9127 at $0.05/M. MI355X is 36% cheaper per token; MI355X delivers 9% more tok/s/chip.
Setting 101 tok/s/user as the target on Kimi K3.1 Pro, GB200 NVL72 produces 5689 tok/s/chip ($0.09 per million tokens) and MI355X produces 4222 ($0.10). GB200 NVL72 is 9% cheaper per token; GB200 NVL72 delivers 35% more tok/s/chip.
At 138 tok/s/user interactivity on Kimi K3.1 Pro, GB200 NVL72 delivers 2582 tok/s/chip at $0.20 per million tokens; MI355X delivers 2743 tok/s/chip at $0.15. MI355X is 32% cheaper per token; MI355X delivers 6% more tok/s/chip at this point. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | GB200 NVL72:8351.6MI355X:9127.3 | GB200 NVL72:5688.7MI355X:4222.3 | GB200 NVL72:2581.8MI355X:2743.1 |
| Cost ($/M tok) | GB200 NVL72:$0.062MI355X:$0.046 | GB200 NVL72:$0.091MI355X:$0.099 | GB200 NVL72:$0.200MI355X:$0.152 |
| tok/s/MW | GB200 NVL72:4466095MI355X:4367151 | GB200 NVL72:3042083MI355X:2020232 | GB200 NVL72:1380646MI355X:1312508 |
| Concurrency | GB200 NVL72:~41MI355X:~16 | GB200 NVL72:~30MI355X:~6 | GB200 NVL72:~12MI355X:~3 |