DeepSeek V4 Pro 1.6T — B200 vs MI300X Performance per Dollar
Cost per million tokens of B200 (NVIDIA Blackwell) versus MI300X (AMD CDNA 3) on DeepSeek V4 Pro 1.6T. 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.
At 61 tok/s/user on DeepSeek V4 Pro 1.6T, B200 comes in at $0.07 per million tokens. MI300X hasn't been benchmarked at this operating point.
Only B200 has cost data at 116 tok/s/user on DeepSeek V4 Pro 1.6T — $0.28 per million tokens. MI300X is unmeasured at this target.
B200 costs $0.79 per million tokens at 171 tok/s/user on DeepSeek V4 Pro 1.6T; we have no MI300X benchmark data at this exact target. (Numbers reflect the default 8k/1k · 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 · MI300X $0.95/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:$0.068MI300X:— | B200:$0.285MI300X:— | B200:$0.791MI300X:— |
| Concurrency | B200:~1778MI300X:— | B200:~146MI300X:— | B200:~31MI300X:— |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.