Have you ever felt a sense of cold sweat or discomfort at the phrase 'do what you love'?Open social media and you are flooded ...

Claude Code Auto Mode security exploit: Johann Rehberger's Python module-shadowing attack achieves remote code execution 60-80 percent of the time against a feature Anthropic marketed with a 0.00 ...

Extracting items from receipts, picking up schedules from school handouts, reading model numbers from labels. These features can now be built by calling a single API. Building them has become easy.

A new phishing kit abuses a legitimate Microsoft device authorization flow intended for use with printers or smart TVs to steal authentication tokens, register attacker-controlled devices and gain ...

就在刚刚，智谱 GLM 首席科学家唐杰在 X 平台分享了一项关于 GLM-5.3-Flash 推理系统优化的研究。 他透露，从 GLM-5.3-Flash 首次运行在国产 AI ...

Axelera's betting enterprises don't need Google-scale power bills to run serious AI—just a smarter chip that fits the servers ...

智谱今天上线 GLM-5.3-FlashX，推理速度最高 200 tokens/s，较 GLM-5.3-Flash 模型 提速 5 倍、提价 2.5 倍，API 同步开放，模型名称：GLM-5.3-FlashX在这背后，是在 10万 ...

IRVINE, Calif. - Chroma Systems Solutions Inc. has added support for Rohde & Schwarz MXO Series digital oscilloscopes to its ATS 8000 Power Conversion Device Automated Test System for automated ...

Z.ai on September 17, 2026 published a technical account describing how it built a complete production-grade inference service for its GLM-5.3-Flash model from scratch on a cluster of more than ...

GLM，已经开始参与构建GLM了。 清华大学计算机系教授、智谱创始人唐杰刚刚分享了一个他们内部观察到的RSI早期案例： GLM-5.3驱动的Infra Agent，在超过10万张国产芯片组成的集群上，从零参与搭建并优化了一套生产级推理系统。 不到两周 ...

Microsoft says threat actors posing as IT helpdesk staff are tricking employees into phishing and device-code attacks that ...

2026年春天，AI圈悄悄掀起一场静默革命——不是又一个‘万亿参数’的吹哨，也不是某家巨头闭源模型的突然空降，而是一群学者把‘老派’自回归（AR）大模型，亲手塞进了一台自带涡轮增压的跑车引擎里。 没错，就是那个被我们用了整整五年、每天都在Chat界面背后吭哧吭哧逐字吐Token的AR架构，终于不用再‘单线程打工’了。 最近，由基础模型研究院（IFM）联合伊利诺伊大学厄巴纳-香槟分校、康奈尔科技学院 ...