AITOPAI热报
ModelsMulti-source

华为发布连接4000颗芯片的AI集群

Huawei unveils AI cluster linking up to 4,000 chips

Tech in Asia@Minh Le
Blog/Media· 4 sources
Read the original
Why we picked it

华为搞了个能连4000颗芯片的AI集群，算力比之前强多了，应该能训练出更厉害的大模型。

华为推出了一款名为昇腾的AI集群，旨在突破美国对半导体相关的限制。该集群能够整合大量计算资源，提升AI模型的训练和推理效率。目前，该集群已成功连接了4000颗芯片，为大型模型训练提供了强大的算力支持。

Original · Tech in Asia

Huawei unveils AI cluster linking up to 4,000 chips

Huawei seeks to build AI systems despite US semiconductor-related restrictions.

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Models" desk, focusing on China's AI industry and domestic foundation model ecosystem. Within our cross-regional analytical framework, it directly maps to US-China foundation model benchmarks and commercialization comparison. Key entities and technical topics under continuous tracking include Huawei, 昇腾, AI集群, 芯片.

Source attribution: curated from Tech in Asia (@Minh Le). Cross-source corroboration: 4 independent sources have covered this development (including IT之家, techcrunch, pandaily), confirming industry-wide signal strength.

#Huawei#昇腾#AI集群#芯片

Keep reading

Recent stories with the same tag or section
IndustryChina to ASEANMulti-source01:11
Chinese contentCP 联合 7 家伙伴举办 Dolphin Tank 决赛路演，展示 7 个 AI 项目

CP 集团拉上 AWS、华为、腾讯云等 7 家伙伴办了个 AI 加速器，7 个项目全从真实业务需求出发，不是纸上谈兵。想知道泰国大企业怎么落地 AI、中美元厂商在东南亚怎么同台，这篇值得看。

#CP-CoEX#Dolphin Tank#Huawei#Tencent Cloud
Event
Techsauce@ Techsauce Team Source
IndustryBuildersSelected12:15
PyTorch Foundation maps Chinese contributions to vLLM and Ascend

The non-obvious signal is that Huawei's Ascend stack is becoming mainstream upstream rather than a parallel ecosystem, lowering the risk of betting on Chinese silicon for inference and training. Framework maintainers and infra teams evaluating Ascend as a CUDA alternative should track TorchNPU and vLLM merge activity as their health metric. The OpenRouter token-share figure also gives a concrete data point for anyone benchmarking open-weight model adoption against US-hosted models.

#PyTorch#Huawei#Ascend#vLLM
Official accountPyTorch@PyTorchSource
IndustryBuilders18:05
Chinese contentAI 基础设施热潮将中国“土豆之都”Ulanqab 变成新超算数据中心枢纽

乌兰察布靠低温和便宜风电把数据中心电费打了下来，DeepSeek、Huawei、Alibaba、Apple 都把算力放这里了。想知道你的推理成本为什么能降、国内算力版图怎么铺的，这篇把选址逻辑讲得挺清楚，做 AI 基建采购的值得看看。

#DeepSeek#Huawei#Alibaba#数据中心
GenkSource
ModelsBuildersOfficial10:49
Chinese contentHuawei openJiuwen 推出 TTSE 智能体框架：将环境事实与条件提示分离

华为 openJiuwen 开源了 TTSE，把智能体经验分成环境事实和条件提示两轨管理，在 ALFWorld 跑到 97.8%，还减少了办公任务的工具调用。做智能体长期记忆和经验积累的可以看看源码，不过数据是自家评测，先自己跑一遍再下结论。

#openJiuwen#Huawei#智能体#ALFWorld
Event
Primary sourcepandaily@[email protected] (Pandaily)Source
ModelsBuildersSelected01:23
BEHAVIOR Challenge at CoRL 2026 crowns Huawei Noah EAI Team

Nearly half of long-horizon household tasks now complete successfully, up from 12% a year earlier — a pace of improvement that suggests embodied agents are approaching practical utility faster than most roadmaps assumed. Teams building home or service robots should treat the BEHAVIOR task suite as a reference evaluation and study the winning approaches. The all-Chinese podium also matters for anyone tracking where embodied AI talent and compute are concentrating.

#BEHAVIOR#Stanford#Huawei#Xiaomi
Event
Official accountFei-Fei Li@drfeifeiSource