DeepSeek开源昇腾版AI基础设施
DeepSeek把训练V4模型的底层工具开源了，现在昇腾芯片也能用上了，和英伟达版本功能一样。
DeepSeek将训练V4模型时开发的整套基础设施开源，包括TileLang编译工具、计算库和分布式通信库。这套基础设施原本运行在英伟达GPU上，现已移植到华为昇腾芯片。
所有组件与英伟达版本一一对应，开发者可直接使用。
刚刚DeepSeek把自家整套昇腾版AI基础设施全部开源
DeepSeek此前为了训练V4系列模型，打磨出了一套底层基础设施：编程语言、矩阵计算库、通信库等，V4训练的大部分算子是基于这套东西实现的
过去这套东西是跑在英伟达GPU上的，现在他们把同一套组件移植到了华为昇腾芯片上，并全部开源
涵盖TileLang高级语言编译工具、计算库、分布式通信库，所有组件与英伟达的一一对应
#DeepSeek
This report is categorized under AITOP's "Products" desk, focusing on China's AI industry and domestic foundation model ecosystem. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows and US-China foundation model benchmarks and commercialization comparison. Key entities and technical topics under continuous tracking include DeepSeek, Ascend, V4, open-source models.
Source attribution: curated from AIGCLINK (@aigclink). Cross-source corroboration: 10 independent sources have covered this development (including Lxfater, IT之家, Gorden Sun), confirming industry-wide signal strength.
