GLM 5.3 744B — TPU7x vs Vera Rubin NVL72
Head-to-head AI inference benchmark comparison of TPU7x (Google Ironwood) and Vera Rubin NVL72 (NVIDIA Vera Rubin) on GLM 5.3 744B. 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 →
TPU7x: 7846 tok/s/chip, $0.04 per million tokens at 92 tok/s/user on GLM 5.3 744B. Vera Rubin NVL72 is unmeasured here.
At 164 tok/s/user on GLM 5.3 744B, TPU7x delivers 5529 tok/s/chip at $0.05 per million tokens; Vera Rubin NVL72 hasn't been benchmarked at this target.
TPU7x hits 3538 tok/s/chip for $0.08 per million tokens at 237 tok/s/user on GLM 5.3 744B. No Vera Rubin NVL72 data at this operating point. (Numbers reflect the default agentic-traces · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | TPU7x:7846.2Vera Rubin NVL72:— | TPU7x:5529.4Vera Rubin NVL72:— | TPU7x:3538.3Vera Rubin NVL72:— |
| Cost ($/M tok) | TPU7x:$0.036Vera Rubin NVL72:— | TPU7x:$0.052Vera Rubin NVL72:— | TPU7x:$0.081Vera Rubin NVL72:— |
| tok/s/MW | TPU7x:6500540Vera Rubin NVL72:— | TPU7x:4581105Vera Rubin NVL72:— | TPU7x:2931520Vera Rubin NVL72:— |
| Concurrency | TPU7x:~12Vera Rubin NVL72:— | TPU7x:~6Vera Rubin NVL72:— | TPU7x:~3Vera Rubin NVL72:— |