Kimi K3 2.8T — GB300 NVL72 vs TPU7x Performance per Dollar
Cost per million tokens of GB300 NVL72 (NVIDIA Blackwell) versus TPU7x (Google Ironwood) on Kimi K3 2.8T. Large-hyperscaler-volume ownership TCO normalized by total 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 →
GB300 NVL72: $0.08 per million tokens. TPU7x: $0.07. Both at 116 tok/s/user on Kimi K3 2.8T, with TPU7x 13% cheaper.
Around the middle of the 26–390 tok/s/user interactivity band — at 208 tok/s/user — GB300 NVL72 runs $0.16 per million tokens on Kimi K3 2.8T while TPU7x runs $0.19. GB300 NVL72 is the cheaper choice by 22%.
Cost-per-million is essentially even between GB300 NVL72 ($0.30) and TPU7x ($0.30) at 299 tok/s/user on Kimi K3 2.8T. (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): GB300 NVL72 $2.31/chip/hr · TPU7x $1.03/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 | GB300 NVL72:$0.081TPU7x:$0.071 | GB300 NVL72:$0.158TPU7x:$0.193 | GB300 NVL72:$0.302TPU7x:$0.304 |
| Concurrency | GB300 NVL72:~47TPU7x:~21 | GB300 NVL72:~4TPU7x:~6 | GB300 NVL72:~1TPU7x:~3 |
Inference Performance
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
No data available
No measurements to plot for this selection. Review the benchmark controls above or adjust quick filters.