智能体落地与模型生态分化
智能体从演示到生产:自主执行与多任务管理
-
Codex Autonomously Completes Security Audit and Earns Bounty
A user instructed the AI model Codex to earn $5, and Codex autonomously completed a series of tasks: finding an…
-
Show HN: Statewright — boosting AI agent reliability with visual state machines
Statewright is a system that constrains AI agents through state machines, controlling which tools they can use at each…
-
Grok接入Gmail,智能邮件助手革新收件箱管理
Grok现已支持连接Gmail,用户可通过自然语言指令对收件箱进行智能查询与管理。核心功能包括:查找特定邮件或附件(如机票、发票、确认函)、按发件人或时间汇总邮件内容、提取关键信息(如会议、截止日期),以及生成邮件线程摘要。该集成旨在将传…
模型生态分化:万亿参数、交互模型与垂直优化
-
AntLingAGI releases trillion-parameter model Ring-2.6-1T
AntLingAGI has released Ring-2.6-1T, its trillion-parameter flagship thinking model, which is free to use via…
-
腾讯混元Hy3预览版发布,专注复杂智能体任务
腾讯混元Hy3预览版已开放早期体验,被描述为该系列最强模型。其核心设计面向现实世界有效性,而非单纯追求基准测试分数。模型具备处理复杂智能体任务的能力,采用256K上下文长度,并融合了快慢思维机制的混合专家架构。该模型基于重建的预训练和强化…
企业级AI落地:金融、法律与基础设施
-
Claude Enters the Legal Industry
Anthropic released more than 20 new MCP connectors and 12 dedicated plugins for the legal industry, deeply integrating…
-
OpenAI launches DeployCo to help enterprises build businesses around intelligence
OpenAI has officially launched DeployCo, a new enterprise deployment company aimed at helping organizations put…
AI能力突破:医疗、数学与材料科学
-
Fields Medalist tests ChatGPT 5.5 Pro: paper-level results in 17 minutes, sounding a red alert for students
Fields Medalist Timothy Gowers tested ChatGPT 5.5 Pro, and the AI independently solved an open additive number theory…
-
材料科学AI多任务模型突破
MatterSim正在拓展AI在材料科学中的应用边界——从更快速的大规模模拟,到全新多任务模型MatterSim-MT,可模拟超越势能面的多种物性。https://msft.it/6017vPamT
安全与伦理:AI的双刃剑效应
-
China's First AI Ghostwriting Recommendation Notes Case Ruled; Court Orders Tool Provider to Pay Platform 100,000 Yuan
China's first unfair competition case involving AI ghostwriting of recommendation notes was ruled by the Hangzhou…
-
Agents execute well but fall short on optimizing user interests
Testing through SocialReasoning Bench found a consistent pattern across models: agents can handle execution tasks, but…