Back to News

Former DeepSeek V1 Author Unveils Open-Source Personal Agent Linger

#open-source#personal-agent#local-llm#memory

Tony Zhang, former DeepSeek V1 author and DeepSeek-VL project lead, publicly unveiled Linger, an open-source personal agent optimized for local LLM deployment on personal computers. Linger can operate existing web, email, and calendar tools without extra credentials, supports third-party models, and runs a 125B-parameter Qwen3.8-Flash-Next on a 12GB RTX 3060 at 45.5 tokens per second. Its core 'Another Me' system builds a local knowledge base from user data and achieves 95% accuracy on the LoCoMo benchmark, surpassing memory-enhanced systems like Mem0 and MemOS.

Coverage timeline

  1. 量子位量子位

    刚刚,DeepSeek V1作者创业项目曝光:开源Agent Jay 2026-11-21 14:32:31 来源: 量子位 专为本地部署而生 聚是一团火,散是满天星。 刚刚,前DeepSeek V1作者、DeepSeek-VL项目负责人 Tony Zhang ,正式公布了自己的新创业项目: Linger 。 方向,正是当下最火的个人 Personal Agent 。 X 帖子视频.mp4 路线也是比较有意思,当同行都在做APP,他们优先选择了电脑。 这意味着,Linger能操纵你电脑已有的网页、邮件箱、日历,无需折腾一大堆登录凭据。 遇到卡住的地方,你也可以直接上手接管。 X 评论视频.mp4 另外一个点,作为中间层的Linger,支持接入各个第三方模型。 并且专门为本地LLM做了优化。 具体来说,一块12GB显存的RTX 3060显卡,就能跑起一个125B参数的 Qwen3.8-Flash-Next ,并且 是45.5 tokens/秒的生成速度。 对了,还有个蛮重要的点—— 这套系统是开源的。。。 至于为什么? 在官宣帖中,Tony Zhang引用了PC崛起的历史: 1960年代,计算是中心化的,直到PC将权力交到每个人手中。如今,AI在数据中心训练,然后以token的形式租给你。 我们相信智能应该是分布式的。你的上下文、你的数据,以及驱动它们的智能,都应该完全由你拥有。 所以今天,他们开源了Linger,一个生活在你个人电脑上的专属Agent。 也是延续大肥鲸优良传统了😭。 一个「会主动,能自学」的AI Linger的的核心,是名为「Another Me」的Agent系统。 它首先是一个记忆中枢。 Linger会在本地读取你的收件箱、日历、聊天记录和会议,每晚自动将这些信息整理成个人知识库 (Wiki) ,供所有Agent查询。 在衡量长期对话记忆能力的LoCoMo基准测试中, Linger的准确率达到了95% ,超过了Mem0、MemOS等多个记忆增强型AI系统。 有了这些,Linger便能更好地主动工作。 它会默默观察你的信息流,根据事件的重要性和紧急性决定何时提醒你。周一收到客户邮件,要求周五前给报价,它不会立刻打扰,而是等到周三早上再弹出提醒。 但如果是会议时间变更,它会立即通知。 这个Agent还会从你的反馈中自我学习,Linger称之为「隐式强

Related stories

  • DeepSeek open-sources Super-agent architecture for autonomous bug-fixing agents

    DeepSeek has publicly released the architectural blueprint for its Super-agent, a research prototype of a personal agent that maintains cross-session context to handle multi-task coordination. Internal tests show the agent can autonomously identify bugs and submit pull requests without being explicitly prompted, with a median task completion time of 12.4 minutes in simulated multi-task environments. DeepSeek positions this as a step toward reducing human attention bottlenecks in multi-task workflows.

    3 reports · 3 sources
  • Ant Group's Bailing open-sources Ling-3.1-flash-1019 for personal agents

    On October 28, 2026, Ant Group's Bailing open-sourced Ling-3.1-flash-1019, an updated model for personal-agent tasks, along with its base model Ling-3.1-flash-base. The new version scores 44 on the Artificial Analysis Intelligence Index v4.3.2, up 3 points from the previous version, and shows improvements across benchmarks including AA-Briefcase, GDPval-AA, AutomationBench-AA, CritPt, and SciCode. The release is linked to the Lingguang 2.0 upgrade, which positions the assistant as a proactive personal AI that can manage long-running tasks.

    1 reports · 1 sources
  • Google and collaborators release CLASHarness benchmark for multi-agent conflict in computer use

    Researchers from Google, Emory, USC, UF, UChicago, UCSB, and Columbia released CLASHarness, a benchmark and pipeline for evaluating 15 major LLMs on multi-agent conflict scenarios in computer-use tasks. The dataset includes 600 multi-turn executable conflict tasks across 20 domains, with 7,970 4K visual evidence images and 3,284 mandatory checkpoints, and is open-sourced on GitHub.

    1 reports · 1 sources
  • Chinese AI startups HyperVolve and Apexin announce funding and model releases

    Chinese AI startup HyperVolve announced tens of millions RMB in angel funding led by Lanchi Ventures, with participation from SenseTime Guoxiang Capital, Lenovo Star, Inno Fund, and Tsinghua alumni seed fund, to build research foundation models using recursive self-improvement. Separately, Apexin, founded by 27-year-old Tsinghua assistant professor Chen Yongchao, released Apex-35B-V1.0, a 35B model claiming SOTA results on 21 benchmarks against comparable models.

    2 reports · 1 sources

News updates every hour. Subscribe in your RSS reader to get new stories automatically.

Subscribe