Kimi K3 2.8T — B200 vs H200 Performance per Dollar
Cost per million tokens of B200 (NVIDIA Blackwell) versus H200 (NVIDIA Hopper) 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 →
Near the low end of the 3–5 tok/s/user interactivity band — at 3 tok/s/user — B200 runs $0.05 per million tokens on Kimi K3 2.8T while H200 runs $2.86. B200 is the cheaper choice by 5308%.
On Kimi K3 2.8T at 4 tok/s/user, the per-million math comes out to $0.05 for B200 and $3.74 for H200; B200 delivers 6966% more output per dollar.
At 5 tok/s/user on Kimi K3 2.8T, B200 costs $0.05 per million tokens; H200 costs $4.68. B200 is 8745% 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): B200 $1.73/chip/hr · H200 $1.22/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:$0.053H200:$2.862 | B200:$0.053H200:$3.740 | B200:$0.053H200:$4.682 |
| Concurrency | B200:~96H200:~7 | B200:~96H200:~5 | B200:~96H200:~3 |
Inference Performance
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.