Yuchen Jin· @Yuchenj_UW · X·· 1 小时前AI 评分62
AI 导读
DeepSeek 官方宣布开源大规模云服务推理框架 DeepInfer，此前已开源 TileLang 及一系列推理算子库。基于该框架为 DeepSeek V4.1 Flash 搭建的高吞吐低延迟推理服务，单集群吞吐达 155 trillion tokens/day，按非高峰价格计算整体收入为每卡每小时 $38。作者 Yuchen Jin 评价称 DeepSeek 连收入数据也一并开源，虽未披露所用 GPU，但 DeepInfer 同时支持 NVIDIA Blackwell 和华为昇腾；即使按 $7/GPU-hour 租用 B300 计算，毛利率也约 82%。
正文 · 原文
“A single cluster currently serves 155 trillion tokens/day, generating $38 in revenue per GPU-hour (at off-peak pricing).”
Huge flex from DeepSeek. Not only did they open-source their inference engine, but they also open-sourced their revenue numbers.
They didn't disclose which GPUs they're using here, but DeepInfer supports both NVIDIA Blackwell and Huawei Ascend.
Even assuming they're renting B300s at $7 per GPU-hour, that's ~82% gross margin.
来源：Yuchen Jin · x.com