gpt-oss 120B — B200 vs MI300X Performance per Dollar
Cost per million tokens of B200 (NVIDIA Blackwell) versus MI300X (AMD CDNA 3) on gpt-oss 120B. 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.
On gpt-oss 120B at 106 tok/s/user, the per-million math comes out to $0.03 for B200 and $0.15 for MI300X; B200 delivers 484% more output per dollar.
At 154 tok/s/user on gpt-oss 120B, B200 costs $0.05 per million tokens; MI300X costs $0.30. B200 is 562% more cost-efficient at this operating point.
B200 edges MI300X at 201 tok/s/user on gpt-oss 120B — $0.07 per million tokens versus $0.80, a 1012% cost-per-token gap. (Numbers reflect the default 1k/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:$0.026MI300X:$0.151 | B200:$0.046MI300X:$0.301 | B200:$0.072MI300X:$0.798 |
| Concurrency | B200:~207MI300X:~17 | B200:~66MI300X:~5 | B200:~64MI300X:~8 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.