👻 Ghostty is a fast, feature-rich, and cross-platform terminal emulator that uses platform-native UI and GPU acceleration.

Understand and test language model architectures on synthetic tasks.

A tool to detect infrastructure issues on cloud native AI systems

Implementation of Rotary Embeddings, from the Roformer paper, in Pytorch

Molecular property prediction for toxicity prediction using GNNs

dstack is a lightweight, open-source alternative to Kubernetes & Slurm, simplifying AI container orchestration with multi-cloud & on-prem support. It natively supports NVIDIA, AMD, & TPU.

TensorDict is a pytorch dedicated tensor container.

Serve, optimize and scale PyTorch models in production

Pytest plugin which splits the test suite to equally sized "sub suites" based on test execution time.

ring-attention experiments

Efficient optimizers

Simple and safe system's rsync wrapper for Python

NanoGPT (124M) in 3.4 minutes

Mamba SSM architecture

Fast and memory-efficient exact attention

LLM training code for Databricks foundation models

A playbook for systematically maximizing the performance of deep learning models.

The simplest, fastest repository for training/finetuning medium-sized GPTs.

codes for "Universal Functional Regression with Neural Operator Flows"

Kolmogorov Arnold Networks

Distributed GPU-Accelerated Framework for Evolutionary Computation. Comprehensive Library of Evolutionary Algorithms & Benchmark Problems.

Schedule-Free Optimization in PyTorch

Fast and flexible physics-based battery models in Python

A set of real-world multi-objective optimization problems

The Wren sits on its Roost in the Aviary.

A materials discovery algorithm geared towards exploring high-performance candidates in new chemical spaces.

Spherical Coordinate system, Rectangular coordinate system in N-sphere (Using Python)