Head-to-head AI inference benchmark comparison of GB200 NVL72 (NVIDIA Blackwell) and GB300 NVL72 (NVIDIA Blackwell) on Qwen 3.5 397B-A17B. 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 190 tok/s/user on Qwen 3.5 397B-A17B: GB200 NVL72 hits 43532 tok/s/chip, GB300 NVL72 hits 102405. Per-million costs land at $0.01 and $0.01 respectively. GB300 NVL72 is 89% cheaper per token; GB300 NVL72 delivers 135% more tok/s/chip.
GB200 NVL72 / GB300 NVL72 on Qwen 3.5 397B-A17B at 299 tok/s/user: 22337 / 65008 tok/s/chip, $0.02 / $0.01 per million tokens. GB300 NVL72 is 134% cheaper per token; GB300 NVL72 delivers 191% more tok/s/chip.
Toward the upper edge of the 81–518 tok/s/user interactivity band, at 409 tok/s/user on Qwen 3.5 397B-A17B: GB200 NVL72 runs 8843 tok/s/chip at $0.06/M tokens, GB300 NVL72 runs 31417 at $0.02/M. GB300 NVL72 is 186% cheaper per token; GB300 NVL72 delivers 255% more tok/s/chip. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | GB200 NVL72:43531.7GB300 NVL72:102404.7 | GB200 NVL72:22336.5GB300 NVL72:65007.8 | GB200 NVL72:8842.7GB300 NVL72:31417.2 |
| Cost ($/M tok) | GB200 NVL72:$0.012GB300 NVL72:$0.006 | GB200 NVL72:$0.023GB300 NVL72:$0.010 | GB200 NVL72:$0.058GB300 NVL72:$0.020 |
| tok/s/MW | GB200 NVL72:23278973GB300 NVL72:48304122 | GB200 NVL72:11944678GB300 NVL72:30664059 | GB200 NVL72:4728736GB300 NVL72:14819457 |
| Concurrency | GB200 NVL72:~24GB300 NVL72:~587 | GB200 NVL72:~13GB300 NVL72:~73 | GB200 NVL72:~6GB300 NVL72:~50 |
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
Qwen3.5 397B 8K / 1K Total Tokens per $1 TCO vs. Interactivity