SemiAnalysis· @SemiAnalysis_ · X·· 20 天前AI 评分47
AI 导读
突发新闻🚨:Google 已将其首个 TPUv7 FP8 vLLM 结果提交至我们的 Agentic Inference Benchmark,并展示了每美元性能比 GB300 NVL72 FP4 还高 1.67 倍。🚀 看到 Google 全力推进将 TPU 的能力对外输出,实在令人惊叹。👇️ (1/4)🧵
AI 生成摘要 · 以原文为准
正文
BREAKING NEWS🚨: Google has submitted its FIRST TPUv7 FP8 vLLM results to our Agentic Inference Benchmark and has demonstrated 1.67x better perf per dollar than even GB300 NVL72 FP4. 🚀
It is amazing to see Google go full steam ahead with externalizing the power of TPUs outside of Google.👇️ (1/4)🧵
来源:SemiAnalysis · x.com