Kimi K3 2.8T — B200 vs GB300 NVL72 Performance per Dollar
Cost per million tokens of B200 (NVIDIA Blackwell) versus GB300 NVL72 (NVIDIA Blackwell) on Kimi K3 2.8T. Owning-hyperscaler TCO normalized by output tokens — performance per dollar across LLM workloads. Pick the more cost-efficient SKU at every target interactivity level. 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: $0.13 per million tokens. GB300 NVL72: $0.08. Both at 56 tok/s/user on Kimi K3 2.8T, with GB300 NVL72 62% cheaper.
Around the middle of the 7–206 tok/s/user interactivity band — at 106 tok/s/user — B200 runs $0.21 per million tokens on Kimi K3 2.8T while GB300 NVL72 runs $0.13. GB300 NVL72 is the cheaper choice by 68%.
On Kimi K3 2.8T at 156 tok/s/user, the per-million math comes out to $0.44 for B200 and $0.19 for GB300 NVL72; GB300 NVL72 delivers 130% more output per dollar. (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.)
Chip pricing (owning hyperscaler): B200 $1.73/chip/hr · GB300 NVL72 $2.31/chip/hr. Source: SemiAnalysis Market July 2026 Pricing Surveys & AI Cloud TCO Model.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Dollar per Million Tokens | B200:$0.131GB300 NVL72:$0.081 | B200:$0.211GB300 NVL72:$0.126 | B200:$0.442GB300 NVL72:$0.192 |
| Concurrency | B200:~12GB300 NVL72:~63 | B200:~6GB300 NVL72:~35 | B200:~2GB300 NVL72:~7 |
Inference Performance
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.