GLM 5.3 744B — B200 vs TPU7x
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and TPU7x (Google Ironwood) 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 →
At 98 tok/s/user interactivity on GLM 5.3 744B, B200 delivers 8517 tok/s/chip at $0.06 per million tokens; TPU7x delivers 7532 tok/s/chip at $0.04. TPU7x is 49% cheaper per token; B200 delivers 13% more tok/s/chip at this point.
B200 posts 6578 tok/s/chip for $0.07 per million tokens at 151 tok/s/user on GLM 5.3 744B; TPU7x posts 5978 tok/s/chip for $0.05. TPU7x is 53% cheaper per token; B200 delivers 10% more tok/s/chip.
Throughput at 203 tok/s/user on GLM 5.3 744B: B200 hits 4221 tok/s/chip, TPU7x hits 4281. Per-million costs land at $0.11 and $0.07 respectively. TPU7x is 70% cheaper per token; TPU7x delivers 1% more tok/s/chip. (Numbers reflect the default agentic-traces · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:8516.9TPU7x:7532.3 | B200:6578.1TPU7x:5978.4 | B200:4220.7TPU7x:4280.5 |
| Cost ($/M tok) | B200:$0.056TPU7x:$0.038 | B200:$0.073TPU7x:$0.048 | B200:$0.114TPU7x:$0.067 |
| tok/s/MW | B200:4980628TPU7x:6240484 | B200:3846816TPU7x:4953076 | B200:2468234TPU7x:3546417 |
| Concurrency | B200:~14TPU7x:~12 | B200:~8TPU7x:~6 | B200:~5TPU7x:~4 |