Microsoft Research

AI agent training

Agent Lightning v1.0: A 3,500-Line Lightweight Agentic RL Framework for Training Agents with Real Harnesses

Agent Lightning v1.0 simplifies reinforcement learning training for AI agents without rebuilding frameworks.

Microsoft Research 1d ago

machine learning, forecasting

Forecasting space weather risks on power grids

Machine learning system predicts power grid damage from space weather 30 to 60 minutes ahead.

Microsoft Research 1w ago

AI for biology

Introducing Quine: An AI research system designed for the complexity of biology

Researchers develop Quine, an AI multimodal system designed to model biological complexity.

Microsoft Research 1w ago

AI research

One year in: How Microsoft Research Asia – Singapore is advancing research, partnership and talent for real-world impact

Microsoft Research Asia Singapore explores how frontier AI research creates real-world value.

Microsoft Research 1w ago

robotics

Offloaded inference for real-world physical AI robotics

Microsoft Research shows offloading AI inference improves robotic task performance.

Microsoft Research 2w ago

AI chemistry

Improving synthesis prediction of small molecules at scale with RetroChimera

RetroChimera AI model accelerates chemical synthesis prediction for molecules at scale.

Microsoft Research 2w ago

medical AI models

GigaPath-Flash and GigaTIME-Flash: Toward population-scale discovery with efficient pathology foundation models

GigaPath-Flash and GigaTIME-Flash reduce computational demands for pathology foundation models while maintaining performance.

Microsoft Research 1mo ago

AI research, computational chemistry

Broadening access to Skala creates a faster path to predictive DFT

Microsoft Research releases Skala 1.1 deep-learning functional with improved accuracy and access.

Microsoft Research 1mo ago

MindTopo reveals VLMs’ spatial reasoning abilities

A path, a fence, a knot. MindTopo sets a new benchmark for testing how AI understands topological relationships and highlights new opportunities to st...

Microsoft Research 1mo ago

Introducing CARE-X: Towards Clinically Useful Radiology VLMs with Auxiliary Supervision, Reward-Aligned Learning, and Tool-Augmented Measurement

Radiology AI is evolving beyond report generation. CARE-X explores a unified approach that combines flexible reasoning, calibrated predictions, and me...

Microsoft Research 1mo ago

Orchard: An open framework for scalable agentic AI

Orchard is an open-source framework for the research community to train and evaluate AI agents across task types. It reduces complexity while supporti...

Microsoft Research 2mos ago

Agent Lightning v1.0: A 3,500-Line Lightweight Agentic RL Framework for Training Agents with Real Harnesses

Forecasting space weather risks on power grids

Introducing Quine: An AI research system designed for the complexity of biology

One year in: How Microsoft Research Asia – Singapore is advancing research, partnership and talent for real-world impact

Offloaded inference for real-world physical AI robotics

Improving synthesis prediction of small molecules at scale with RetroChimera

GigaPath-Flash and GigaTIME-Flash: Toward population-scale discovery with efficient pathology foundation models

Broadening access to Skala creates a faster path to predictive DFT

MindTopo reveals VLMs’ spatial reasoning abilities

Introducing CARE-X: Towards Clinically Useful Radiology VLMs with Auxiliary Supervision, Reward-Aligned Learning, and Tool-Augmented Measurement

Orchard: An open framework for scalable agentic AI

Latest AI News

Choose sides Choose whose side you're on

Install it as an app

Ready for more?

Spread the word

PWA successfully installed.
Take that, Apple! 😉

Agent Lightning v1.0: A 3,500-Line Lightweight Agentic RL Framework for Training Agents with Real Harnesses

Forecasting space weather risks on power grids

Introducing Quine: An AI research system designed for the complexity of biology

One year in: How Microsoft Research Asia – Singapore is advancing research, partnership and talent for real-world impact

Offloaded inference for real-world physical AI robotics

Improving synthesis prediction of small molecules at scale with RetroChimera

GigaPath-Flash and GigaTIME-Flash: Toward population-scale discovery with efficient pathology foundation models

Broadening access to Skala creates a faster path to predictive DFT

MindTopo reveals VLMs’ spatial reasoning abilities

Introducing CARE-X: Towards Clinically Useful Radiology VLMs with Auxiliary Supervision, Reward-Aligned Learning, and Tool-Augmented Measurement

Orchard: An open framework for scalable agentic AI

Introducing CARE-X: Towards Clinically Useful Radiology VLMs with Auxiliary Supervision, Reward-Aligned Learning, and Tool-Augmented Measurement