DeepSeek R1 — B200 vs B300
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and B300 (NVIDIA Blackwell) on DeepSeek R1. Latency, throughput, and cost across LLM workloads. Use the chart controls below to switch sequences, precisions, and metrics — same interactions as the main inference chart.
B200 posts 2804 tok/s/chip for $0.17 per million tokens at 72 tok/s/user on DeepSeek R1; B300 posts 3887 tok/s/chip for $0.16. B300 is 6% cheaper per token; B300 delivers 39% more tok/s/chip.
Throughput at 135 tok/s/user on DeepSeek R1: B200 hits 1008 tok/s/chip, B300 hits 921. Per-million costs land at $0.48 and $0.68 respectively. B200 is 43% cheaper per token; B200 delivers 9% more tok/s/chip.
B200 / B300 on DeepSeek R1 at 198 tok/s/user: 373 / 654 tok/s/chip, $1.29 / $0.96 per million tokens. B300 is 34% cheaper per token; B300 delivers 75% more tok/s/chip. (Numbers reflect the default 8k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:2804.4B300:3886.8 | B200:1008.5B300:921.3 | B200:373.2B300:653.7 |
| Cost ($/M tok) | B200:$0.171B300:$0.162 | B200:$0.477B300:$0.681 | B200:$1.288B300:$0.960 |
| tok/s/MW | B200:1639990B300:2045668 | B200:589747B300:484897 | B200:218222B300:344074 |
| Concurrency | B200:~208B300:~95 | B200:~14B300:~6 | B200:~2B300:~17 |
Inference Performance
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
Total Tokens per $1 USD (Owning - Hyperscaler) vs. Interactivity
DeepSeek R1 0528 671B • FP4 • 8K / 1K • Source: SemiAnalysis InferenceX™
TCO $/chip/hr: H100: 1.17H200: 1.22B200: 1.73B300: 2.26GB200: 1.86GB300: 2.31MI300X: 0.95MI325X: 1.1MI355X: 1.5RTX6000PRO: 0.68
Source: SemiAnalysis Market July 2026 Pricing Surveys & AI Cloud TCO Model
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate cost per million tokens per decode chip or per prefill chip, rather than per total chip count. This makes direct token cost comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate input throughput per decode chip or per prefill chip, rather than per total chip count. This makes direct input throughput comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate output throughput per decode chip or per prefill chip, rather than per total chip count. This makes direct output throughput comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate power per decode chip or per prefill chip, rather than per total chip count. This makes direct power comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate Joules per decode chip or per prefill chip, rather than per total chip count. This makes direct Joules per token comparison with aggregated configs not an apples-to-apples comparison.
No data available
Please change the model, sequence, precision, date range or chip selection.
Shift+Scroll to zoom • Drag to pan • Double-click to reset • Click a point to pin tooltip