最新资讯
www.infoq.com GitLab 19.5 将 Agentic 开发纳入 DevOps 平台的治理体系 dev.to 在信任 AI 生成的测试用例前,如何逐条复核 unite.ai Franklin Templeton 构建 AI 研究工作流以揭示风险 techxplore.com AI 助力寻找更优的钛合金 3D 打印工艺参数 engadget.com Perplexity 推出“teammates”日常工作任务 AI agent computerworld.com 别提数据中心:OpenAI 改称“compute campuses” theguardian.com 俄美试图阻止就自主武器制定具有约束力的规则 bloomberg.com Nvidia 加强东南亚芯片追踪以打击走私 thestack.technology 模型路由需求升温,Agent Router 推出开源替代方案 www.youtube.com Google DeepMind 的 Philipp Schmid:AI 同事需要具备的七个要素 unite.ai Cashfree 为 AI Agent 的托管结账加入 WebMCP 工具 dev.to 我们在 DolphinBench 上测试记忆检索:纯关键词搜索暴露了自身回归
多元动态
登录 注册
大模型 www.aibase.com ·

阿里平头哥发布AI芯片真武V900,算力达M890三倍

原文 我的收藏

在2026云栖大会上,阿里平头哥发布国产AI芯片真武V900,宣称算力为前代真武M890的三倍,基于该芯片的单集群可扩展至50万卡,面向大模型训练与推理。阿里同时表示M890 AI超节点已具备支持2万亿参数大模型推理的能力,并已在本季度于阿里云数据中心上线,此外开源Qwen 2.7B模型并推出Qwen移动端AI方案。上述算力与规模数据均为厂商宣称,缺乏技术细节与独立验证。

AI推荐理由 内容涉及阿里平头哥在2026云栖大会发布AI芯片真武V900,宣称算力为前代M890三倍,单集群可扩展至50万卡,并提及M890超节点支持2万亿参数模型推理、开源Qwen 2.7B及Qwen移动端方案。信息密度较高且覆盖芯片、集群、云基础设施和端侧生态,但来源为AIbase聚合转述,非官方一手发布,正文中'Jack Ma, CEO of Alibaba Group'与事实不符(马云并非阿里集团CEO),削弱可信度;算力三倍、50万卡、2万亿参数等关键数据均无技术细节或第三方验证,属厂商宣称,需保留不确定性。

62

正文

Tongyi Pingtouge Launches AI Chip Zhenwu V900 with Three Times the Computing Power of M890 AIbase基地 Published inAI News · 2 min read · Sep 22, 2026 10 At the 2026 Yunqi Conference, Alibaba's T-Head released the domestic AI chip Zhenwu V900, with computing power three times that of its predecessor Zhenwu M890. A single cluster built based on Zhenwu V900 can be scaled up to 500,000 cards, providing large-scale computing power support for cutting-edge AI model training and inference. Alibaba stated that as the T-Head chip product line becomes increasingly mature and widely adopted by customers, the annual chip shipment volume is expected to increase significantly. Jack Ma, CEO of Alibaba Group, revealed that Alibaba's self-developed M890 AI super node has already gained the capability to support the inference of a 2-trillion parameter large model, and has been officially launched in Alibaba Cloud data centers this quarter. While continuously advancing chip and cloud infrastructure, Alibaba has also further expanded its edge-side AI ecosystem, open-sourcing the Qwen 2.7B model and officially launching the Qwen mobile AI solution. From Zhenwu V900 to M890 AI super node, Alibaba is simultaneously advancing the construction of AI chips, ultra-large scale clusters, cloud infrastructure, and edge-side model ecosystems, covering different computing power needs of cutting-edge models from training, inference to terminal deployment. Zhenwu V900M890 AI Super NodeQianwen 2.7B ModelAlibaba Cloud This article is from AIbase Daily Scan to view Welcome to the [AI Daily] column! This is your daily guide to exploring the world of artificial intelligence. Every day, we present you with hot topics in the AI field, focusing on developers, helping you understand technical trends, and learning about innovative AI product applications. —— Created by the AIbase Daily Team © Copyright AIbase Base 2024, Click to View Source -

来源报道 · 1 篇 · 以下链接均前往原始媒体

← 返回精选动态 原文

更多内容