Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and GB300 NVL72 (NVIDIA Blackwell) on DeepSeekv4 Pro 0813 1.6T. 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.
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 →
Throughput at 96 tok/s/user on DeepSeekv4 Pro 0813 1.6T: B200 hits 42233 tok/s/chip, GB300 NVL72 hits 63340. Per-million costs land at $0.01 and $0.01 respectively. GB300 NVL72 is 12% cheaper per token; GB300 NVL72 delivers 50% more tok/s/chip.
B200 / GB300 NVL72 on DeepSeekv4 Pro 0813 1.6T at 159 tok/s/user: 7947 / 10284 tok/s/chip, $0.06 / $0.06 per million tokens. B200 is 3% cheaper per token; GB300 NVL72 delivers 29% more tok/s/chip.
Toward the upper edge of the 33–284 tok/s/user interactivity band, at 222 tok/s/user on DeepSeekv4 Pro 0813 1.6T: B200 runs 3996 tok/s/chip at $0.12/M tokens, GB300 NVL72 runs 4773 at $0.13/M. B200 is 12% cheaper per token; GB300 NVL72 delivers 19% more tok/s/chip. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:42233.0GB300 NVL72:63340.3 | B200:7946.8GB300 NVL72:10284.4 | B200:3996.5GB300 NVL72:4772.7 |
| Cost ($/M tok) | B200:$0.011GB300 NVL72:$0.010 | B200:$0.060GB300 NVL72:$0.062 | B200:$0.120GB300 NVL72:$0.134 |
| tok/s/MW | B200:24697665GB300 NVL72:29877491 | B200:4647252GB300 NVL72:4851140 | B200:2337107GB300 NVL72:2251296 |
| Concurrency | B200:~175GB300 NVL72:~546 | B200:~10GB300 NVL72:~7 | B200:~4GB300 NVL72:~5 |
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
TCO $/chip/hr:
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.
Matching measurements exist, but their chip series are hidden.
Shift+Scroll to zoom • Drag to pan • Double-click to reset • Click a point to pin tooltip
DeepSeek V4 Pro 0813 1.6T 8K / 1K Total Tokens per $1 TCO vs. Interactivity