Meta Lingua: a lean, efficient, and easy-to-hack codebase to research LLMs.

The code used to train and run inference with the ColPali architecture.

Pytorch implementation of KFAC and E-KFAC (Natural Gradient).

Official open source repository for "A Timing Engine Inspired Graph Neural Network Model for Pre-Routing Slack Prediction" (DAC 2022)

bloop is a fast code search engine written in Rust.

IBM Analog Hardware Acceleration Kit

Fuji is an AI agent that lives in your browser's sidepanel. You can now get tasks done online with a single command!

Tensor Network Library with Autograd

LoRA for arbitrary JAX models and functions

Code to simulate energy-based neuromorphic platforms and learning algorithms for such platforms

DSPy: The framework for programming—not prompting—foundation models

Second Order Optimization and Curvature Estimation with K-FAC in JAX.

Exact OU processes with JAX

Uncertainty quantification with PyTorch

Video and image modeling in the continuous-time (CT) domain.

GradDFT is a JAX-based library enabling the differentiable design and experimentation of exchange-correlation functionals using machine learning techniques.

Structured Text Generation

Compute FID scores with PyTorch.

code for the paper "DiGress: Discrete Denoising diffusion for graph generation"

High-Resolution Image Synthesis with Latent Diffusion Models

Machine learning algorithms for many-body quantum systems