🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Introducing Phi-4-reasoning, adding reasoning models to the Phi family of SLMs. The model is trained with both supervised finetuning (using a carefully curated dataset of reasoning demonstration) and Reinforcement Learning. 📌Competitive results on reasoning benchmarks with

Excited to share our latest Phi model, Phi4-reasoning, a small but powerful model that match the performance of much larger reasoning models up to DeepSeek R1. Here is the report for new insights into training reasoning models and evaluating them: lnkd.in/g_Pz5JQA

Two colleagues recently used our 14-billion parameters Phi-4-reasoning model to ace graduate-level Linear Algebra and Calculus BC tests—scoring 100% and 69/70 respectively. Thanks to the amazing work of our Windows + Devices colleagues, this model now runs on-device on

Magentic-UI is now available via Azure AI Foundry Labs msft.it/6014SjYg4.

In the realm of AI agents for doing complex tasks that require multi-step planning and browser use, Magentic-UI from Microsoft Research is at the cutting edge. I think you'll find it surprisingly useful. Now available in open source on Azure Foundry.

A few months back, our team released Magentic-one -- showing how we can build multi-agent systems with AutoGen for complex web task completion. But how should humans interact with such systems? Magentic-UI shows how to build an agentic user experience, prioritizing

🤖AI agents are getting better. 🙋‍♀️Human agency should too. Today, we're open-sourcing Magentic-UI, an experimental app built on #AutoGen to accelerate research in transparency, control, and human oversight of agentic systems. More 👇

We're pursuing a long-term vision for how AI can amplify human intellect and elevate decision-making in some of the most challenging problems in patient care: questions & directions that arise in tumor board meetings. More in my LinkedIn article: aka.ms/AAwau32

Was a fantastic collaboration on bringing guidance to the full family of OpenAI models. The most comprehensive structured outputs meet the world's best models 🫶 github.com/guidance-ai Shoutout Michal Moskal Andrew Braunstein cc Michelle Pokrass Nikunj Handa Eric Horvitz Kevin Scott

Excited to release my first lead project Magentic-UI at Microsoft Research, an OS web agent application designed for efficient human-agent interaction. CUA agents are cool but they're not so useful yet, Magentic-UI helps us study how to get value from them. github.com/microsoft/mage…

It’s been a pleasure collaborating with Shrey Jain & colleagues at Microsoft’s Health Care & Life Sciences and Microsoft Research on directions ahead for leveraging AI advances to help with cancer care. Stanford Health Care ARPA-H

1K on Magentic-UI ⭐️ Thank you! 🙏 star-history.com/#microsoft/mag… github.com/microsoft/mage… #starhistory #GitHub #OpenSource via Star History

Magentic-UI + Ollama We are slowly adding more support for local models in our new open-source, human-centered browser use agent.

Crazy that it's been almost a decade since my last internship... Super excited to be at Microsoft Research this summer! Will hopefully build an awesome new agentic system with Gagan Bansal and Hussein Mozannar

Check out this new tutorial about a human centered, web agent that we just released!

My recent talk on challenges in developing human-centered agents is now available online! It provides an HCI perspective of our learning from developing AutoGen youtube.com/watch?v=O5jSX8…

What if LLMs could learn your habits and preferences well enough (across any context!) to anticipate your needs? In a new paper, we present the General User Model (GUM): a model of you built from just your everyday computer use. 🧵

The challenge of achieving complementary performance strikes again! h/t Adam Fourney Here's the same exact problem we talked about in the context of XAI a few years. and even with LLM, the same pattern continues to repeat!