implementations and experimentation on mHC by deepseek - https://arxiv.org/abs/2512.24880

Official repository for the Boltz biomolecular interaction models

Code and data accompanying the paper "Quantum Approximate Multi-Objective Optimization"

Optimal transport for evaluating geospatial predictions

Generative AI pipeline for target-based molecular design

PyPI package for number token loss.

Ski lift detector for your Strava activities.

This repository contains sources about reinforcement learning human feedback for math reasoning,.

This code includes the implementation of a Transformer-based Reinforcement Learning approach for Hyper-parameter Optimization

PyMed is a Python library that provides access to PubMed. This is a user-maintained fork from the archived original.

A hierarchical LM that scales to training on context windows of +5M tokens

Pip compatible CodeBLEU metric implementation available for linux/macos/win

Optimal transport tools implemented with the JAX framework, to solve large scale matching problems of any flavor.

Code for BLT research paper

Get your documents ready for gen AI

Addition to the Conditional Monge Gap to extend it to Chimeric Antigen Receptor T cell scRNAseq data

Learning conditional optimal transport maps

This is the code accompanying our ICML 2024 paper "Eureka-Moments in Transformers: Multi-Step Tasks Reveal Softmax Induced Optimization Problems".

Implementation of MEGABYTE, Predicting Million-byte Sequences with Multiscale Transformers, in Pytorch

A regression-alike loss to improve numerical reasoning in language models - ICML 2025

InstructLab Core package. Use this to chat with a model and execute the InstructLab workflow to train a model using custom taxonomy data.

An extremely fast Python package and project manager, written in Rust.

Implementation of Alphafold 3 from Google Deepmind in Pytorch

🏆 A ranked list of awesome python developer tools and libraries. Updated weekly.

🙌 OpenHands: AI-Driven Development