🔎Agentic search like Deep Research is fundamentally changing web search, but it also brings an evaluation crisis⚠️ Introducing Mind2Web 2: Evaluating Agentic Search with Agents-as-a-Judge - 130 tasks (each requiring avg. 100+ webpages) from 1,000+ hours of expert labor -

Zhipu AI & Tsinghua University just unveiled GLM-4.1V-Thinking on Hugging Face! This new 9B VLM leverages scalable RL for versatile multimodal reasoning, matching or exceeding GPT-4o and 72B models on tough benchmarks like STEM & long document understanding.

We are excited to introduce GLM-4.1V-9B-Thinking! Our new open-source model is designed to tackle complex reasoning across a variety of data types, including text, images, and even video. Ready to dive in? Try it out instantly: huggingface.co/spaces/THUDM/G…

GLM-4.1V-Thinking Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning

Introducing Shortcut — the first superhuman Excel agent. Shortcut one-shots most knowledge work tasks on Excel. It even scores >80% on Excel World Championship Cases in ~10 minutes. That's 10x faster than humans. Our early preview is live. Just comment for an invite code.

Come join us at ICML in Vancouver!

Threats in LLM-Powered AI Agents Workflows Neat survey of typical threats you encounter when building AI agents. Prompt injections and protocol exploits included. Bookmark this one!

We're excited to announce Autotab 1.0. The first AI agent that you onboard, not integrate. Models have the intelligence required for most work—but they’re limited by context, tools, and coherence over long tasks. We’ve built Autotab to unhobble models on these bottlenecks 🧵