Kimi K3 2.8T — GB300 NVL72 vs MI355X Performance per Dollar
Cost per million tokens of GB300 NVL72 (NVIDIA Blackwell) versus MI355X (AMD CDNA 4) 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 →
Near the low end of the 19–163 tok/s/user interactivity band — at 55 tok/s/user — GB300 NVL72 runs $0.04 per million tokens on Kimi K3 2.8T while MI355X runs $0.05. GB300 NVL72 is the cheaper choice by 18%.
On Kimi K3 2.8T at 91 tok/s/user, the per-million math comes out to $0.06 for GB300 NVL72 and $0.06 for MI355X; GB300 NVL72 delivers 3% more total tokens per dollar.
At 127 tok/s/user on Kimi K3 2.8T, GB300 NVL72 costs $0.08 per million tokens; MI355X costs $0.14. GB300 NVL72 is 77% more cost-efficient at this operating 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.)
Chip pricing (owning hyperscaler): GB300 NVL72 $2.31/chip/hr · MI355X $1.50/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:$0.042MI355X:$0.050 | GB300 NVL72:$0.063MI355X:$0.065 | GB300 NVL72:$0.081MI355X:$0.143 |
| Concurrency | GB300 NVL72:~56MI355X:~18 | GB300 NVL72:~48MI355X:~30 | GB300 NVL72:~46MI355X:~4 |
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.