DeepSeek V5 plans to pretrain on Huawei Ascend 950 chips
English text is being prepared.
According to reports, DeepSeek's next-generation large model, DeepSeek V5, plans to conduct pretraining on Huawei Ascend chips. Previously, Ascend chips were mainly used for inference tasks in DeepSeek large models; V5 will be its first large model to natively rely on Ascend 950 for training. The Ascend 950 series is divided into two models, 950PR and 950DT, both based on the third-generation Da Vinci architecture. The 950DT is equipped with 144GB of self-developed HBM and a memory access bandwidth of 4TB/s, while the 950PR is equipped with 128GB of on-chip memory and FP4 compute power reaching 1784 TFLOPS. People familiar with the matter said V5 plans to use Ascend 950 as its core training compute, with the project timeline pointing to launch next year.
Why read this
AI score 79
The report outlines the path of Ascend from the inference side to the pretraining side, so readers can judge the actual position of domestic compute in large-model training.
Topics
Body
11月3日消息，据报道，DeepSeek下一代大模型DeepSeek V5计划基于华为昇腾芯片进行预训练。此前华为昇腾芯片主要用于DeepSeek大模型的推理任务，V5将是其首款原生依托昇腾950开展训练的大模型。
华为昇腾950系列分为950PR与950DT两款芯片，均基于第三代达芬奇架构，主要面向大模型的训练与推理场景。
其中，950DT专为训练与推理Decode设计，配备144GB自研HBM，内存访问带宽达4TB/s，互联带宽2TB/s，并原生支持FP8、MXFP8、MXFP4等低精度格式，MXFP4算力较上代BF16提升4倍。
950PR则聚焦推理Prefill阶段，配备128GB片上内存，带宽1.6TB/s，FP4算力达1784 TFLOPS，侧重高吞吐与成本控制。
回顾双方合作历程，昇腾承载DeepSeek的最早规模化部署发生在2025年春节前后。除夕夜到大年初二，华为与DeepSeek完成了3000多张昇腾芯片的部署和验证，将DeepSeek-V3/R1大模型成功部署到华为云贵安数据中心。
2026年4月DeepSeek-V4发布后，华为通过双方芯模技术协同，实现昇腾超节点全系列产品全面支持。DeepSeek当时透露，受限于高端算力，V4-Pro服务吞吐有限，预计昇腾950超节点批量上市后Pro价格会大幅下调。
9月30日，DeepSeek进一步开源面向昇腾的基础设施组件，涵盖TileLang编译工具、计算库与通信库，与英伟达平台开源组件一一对应，将适配延伸到了训练和推理的基础设施底层。
当前国产算力的落地主要集中在推理侧，在训练侧仍需提升。此前虽有团队依托昇腾910C完成DeepSeek-V4-Pro的全参数后训练，但尚未涉及基础预训练。
若V5基于昇腾完成预训练，将首次正面回答中国顶尖大模型能否在国产算力上完成预训练这一核心问题。
据知情人士透露，V5计划采用华为昇腾950芯片作为核心训练算力，项目节奏指向明年落地。若V5顺利基于昇腾950完成训练，将标志着国产AI算力正式迈入支撑万亿级大模型预训练的新阶段。