DeepSeek 刚融资 70 亿美元 — 为何它急需更多资金
deepseek 正在以 710 亿美元估值寻求新一轮融资，距其 70 亿美元融资仅数周，旨在建设数据中心和购买芯片。这笔资金旨在维持其 v4-pro 模型超低价格策略，该价格比竞争对手便宜 11 倍。
deepseek is seeking fresh capital at a $71b valuation to finance proprietary data centers, purchase ai chips, and develop a custom inference chip to reduce reliance on nvidia and huawei. the funding push follows a $7b round closed in late may and is driven by the success of its v4-pro and v4-flash models, whose permanent low prices are driving rapid adoption among us firms. despite trailing western models like openai's gpt-5.6 sol in performance, deepseek's price gap—roughly 11x cheaper on input—has made it a top spending category for tracked businesses. the move signals an intensifying arms race among chinese ai labs, as rivals like zhipu ai and moonshot ai also secure capital to compete on both price and scale.