Muse Spark Open 816B — GB300 NVL72 vs TPU7x
Head-to-head AI inference benchmark comparison of GB300 NVL72 (NVIDIA Blackwell) and TPU7x (Google Ironwood) on Muse Spark Open 816B. 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 179 tok/s/user on Muse Spark Open 816B: GB300 NVL72 hits 35267 tok/s/chip, TPU7x hits 17723. Per-million costs land at $0.02 and $0.02 respectively. TPU7x is 13% cheaper per token; GB300 NVL72 delivers 99% more tok/s/chip.
GB300 NVL72 / TPU7x on Muse Spark Open 816B at 264 tok/s/user: 18269 / 14152 tok/s/chip, $0.04 / $0.02 per million tokens. TPU7x is 74% cheaper per token; GB300 NVL72 delivers 29% more tok/s/chip.
Toward the upper edge of the 95–432 tok/s/user interactivity band, at 348 tok/s/user on Muse Spark Open 816B: GB300 NVL72 runs 13376 tok/s/chip at $0.05/M tokens, TPU7x runs 9123 at $0.03/M. TPU7x is 53% cheaper per token; GB300 NVL72 delivers 47% more tok/s/chip. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | GB300 NVL72:35266.7TPU7x:17722.9 | GB300 NVL72:18268.6TPU7x:14151.8 | GB300 NVL72:13375.7TPU7x:9122.6 |
| Cost ($/M tok) | GB300 NVL72:$0.018TPU7x:$0.016 | GB300 NVL72:$0.035TPU7x:$0.020 | GB300 NVL72:$0.048TPU7x:$0.031 |
| tok/s/MW | GB300 NVL72:16635216TPU7x:14683445 | GB300 NVL72:8617254TPU7x:11724803 | GB300 NVL72:6309311TPU7x:7558056 |
| Concurrency | GB300 NVL72:~21TPU7x:~11 | GB300 NVL72:~11TPU7x:~9 | GB300 NVL72:~8TPU7x:~5 |