Kimi K3 2.8T — B300 vs GB200 NVL72
Head-to-head AI inference benchmark comparison of B300 (NVIDIA Blackwell) and GB200 NVL72 (NVIDIA Blackwell) on Kimi K3 2.8T. 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 25–202 tok/s/user interactivity band, at 69 tok/s/user on Kimi K3 2.8T: B300 runs 4898 tok/s/chip at $0.13/M tokens, GB200 NVL72 runs 7658 at $0.07/M. GB200 NVL72 is 90% cheaper per token; GB200 NVL72 delivers 56% more tok/s/chip.
Setting 114 tok/s/user as the target on Kimi K3 2.8T, B300 produces 3550 tok/s/chip ($0.18 per million tokens) and GB200 NVL72 produces 3363 ($0.15). GB200 NVL72 is 15% cheaper per token; B300 delivers 6% more tok/s/chip.
At 158 tok/s/user interactivity on Kimi K3 2.8T, B300 delivers 2063 tok/s/chip at $0.30 per million tokens; GB200 NVL72 delivers 2695 tok/s/chip at $0.19. GB200 NVL72 is 59% cheaper per token; GB200 NVL72 delivers 31% more tok/s/chip at this point. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B300:4898.1GB200 NVL72:7658.0 | B300:3549.6GB200 NVL72:3362.7 | B300:2063.2GB200 NVL72:2694.9 |
| Cost ($/M tok) | B300:$0.128GB200 NVL72:$0.067 | B300:$0.177GB200 NVL72:$0.154 | B300:$0.304GB200 NVL72:$0.192 |
| tok/s/MW | B300:2577921GB200 NVL72:4095192 | B300:1868235GB200 NVL72:1798243 | B300:1085910GB200 NVL72:1441115 |
| Concurrency | B300:~7GB200 NVL72:~35 | B300:~5GB200 NVL72:~10 | B300:~1GB200 NVL72:~12 |