MiroThinker is an open source deep research agent optimized for research and prediction. It achieves a 60.2% Avg@8 score on the challenging GAIA benchmark.

A research prototype of a human-centered web agent

A partially latent flow matching model for the joint generation of a protein’s amino acid sequence and full atomistic structure, including both the backbone and side chain.

Proteina is a new large-scale flow-based protein backbone generator that utilizes hierarchical fold class labels for conditioning and relies on a tailored scalable transformer architecture.

RNAPro is a state-of-the-art RNA 3D folding model developed in collaboration with the hosts and winners of the Stanford RNA 3D Folding Kaggle competition.

BioNeMo Framework: For building and adapting AI models in drug discovery at scale

A family of codon-resolution language models trained on 130 million protein-coding sequences from over 20,000 species.

Official Python inference and LoRA trainer package for the LTX-2 audio–video generative model.

<msc-ai-prompt /> is a web component based on Chrome Built-in AI Prompt API. Web developers could use <msc-ai-prompt /> wrap form element and give it prompt feature support.

Leaderboard Comparing LLM Performance at Producing Hallucinations when Summarizing Short Documents

This repository is deprecated and will be archived

GitHub Copilot App Modernization

reference server for networkQuality/gorespsonsiveness

A place to share code and server configurations in support of the networkQuality tool

[ICLR 2025] BRIGHT: A Realistic and Challenging Benchmark for Reasoning-Intensive Retrieval

Complex Reasoning Rag System, Agentic Rag System

DDSP: Differentiable Digital Signal Processing

Code for the ICML 2023 paper "SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot".

The code used to train and run inference with the ColVision models, e.g. ColPali, ColQwen2, and ColSmol.

Ongoing research training transformer models at scale

PyTorch emulation library for Microscaling (MX)-compatible data formats

PyTorch native quantization and sparsity for training and inference

Core ML tools contain supporting tools for Core ML model conversion, editing, and validation.