DeepSeek 在知乎披露 V4.1 Flash 推理服务利润率细节
DeepSeek 在知乎自曝家底：V4.1 Flash 每卡每小时收入 $38，有人算出毛利率约 80%，这数据 Twitter 上根本没有。
DeepSeek 在知乎页面披露了 DeepSeek V4.1 Flash 大规模推理服务的运营数据：单个集群日吞吐量达 155 万亿 tokens，按错峰价格计算每张卡每小时收入 $38。推文作者 swyx 估算 GPU 成本约为每小时 $5-10，据此推算 DeepSeek 在自有定价下的服务毛利约 80%。作者同时指出，这类细节只发在知乎，只看 Twitter 的用户看不到这些信息。
deepseek posted way more alpha on their zhihu page, if you only read twitter you are reading the laowai approved stuff
"Based on this framework, we have implemented a high-throughput, low-latency large-scale inference service for DeepSeek V4.1 Flash. Currently, the service throughput of a single cluster is 155 trillion tokens/day, and the overall revenue is $38 per card per hour (based on off-peak prices)."
it costs roughtly $5-10 per gpu hour so depseek is claiming 80% serving margin at THEIR prices lololololol
