Head-to-head AI inference benchmark comparison of GB300 NVL72 (NVIDIA Blackwell) and RTX PRO 6000 (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 →
At 233 tok/s/user on Qwen 3.5 397B-A17B, GB300 NVL72 delivers 108434 tok/s/chip at $0.01 per million tokens; RTX PRO 6000 hasn't been benchmarked at this target.
GB300 NVL72 hits 57224 tok/s/chip for $0.01 per million tokens at 385 tok/s/user on Qwen 3.5 397B-A17B. No RTX PRO 6000 data at this operating point.
GB300 NVL72: 22121 tok/s/chip, $0.03 per million tokens at 537 tok/s/user on Qwen 3.5 397B-A17B. RTX PRO 6000 is unmeasured here. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | GB300 NVL72:108433.9RTX PRO 6000:— | GB300 NVL72:57223.6RTX PRO 6000:— | GB300 NVL72:22120.9RTX PRO 6000:— |
| Cost ($/M tok) | GB300 NVL72:$0.006RTX PRO 6000:— | GB300 NVL72:$0.011RTX PRO 6000:— | GB300 NVL72:$0.029RTX PRO 6000:— |
| tok/s/MW | GB300 NVL72:51148085RTX PRO 6000:— | GB300 NVL72:26992244RTX PRO 6000:— | GB300 NVL72:10434387RTX PRO 6000:— |
| Concurrency | GB300 NVL72:~601RTX PRO 6000:— | GB300 NVL72:~44RTX PRO 6000:— | GB300 NVL72:~120RTX PRO 6000:— |
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. Median Interactivity