First TPUv7 agentic benchmark on vLLM claims 1.42x perf-per-dollar over GB300 NVL72
vllm_project · x · 2026-10-16
SemiAnalysis reported that Google submitted its first TPUv7 result to the AgentX agentic benchmark, served with vLLM (thanks to joint work from Google TPU, RedHat AI, and inferact teams). On an external non-Google workload cost basis, TPUv7 reportedly delivers 1.42x better perf-per-dollar than GB300 NVL72 FP4, and 2.5x better than AMD MI355X. The vLLM project confirmed the milestone and said it expects more models landing on TPU via AgentX.
More from Infra
- First AA-AgentPerf results: NVIDIA Vera Rubin NVL72 delivers up to 5.7x throughput per MW vs GB300 — ArtificialAnlys · 2026-10-16
- Vera Rubin NVL72 first benchmarks: up to 5.7x throughput per MW vs GB300 — ArtificialAnlys · 2026-10-16
- NVIDIA Vera Rubin NVL72 hits up to 5.7x GB300 throughput per MW on agent benchmark — ArtificialAnlys · 2026-10-16
- 14 open-source LLM observability tools to know in 2026, from Langfuse to Laminar — TheTuringPost · 2026-10-16
- Local AI is a routing problem, not a privacy feature: where each task should run — bigdata · 2026-10-16
- Moving agent work from GPUs to CPUs for durable, predictable actions — cramforce · 2026-10-16