The official repository for tariff

Single File, Single GPU, From Scratch, Efficient, Full Parameter Tuning library for "RL for LLMs"

Pretraining code for a large-scale depth-recurrent language model

A PyTorch library for implementing flow matching algorithms, featuring continuous and discrete flow matching implementations. It includes practical examples for both text and image modalities.

The Open Cookbook for Top-Tier Code Large Language Model

Inference script for Oasis 500M

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

[ICLR 2025] Pyramidal Flow Matching for Efficient Video Generative Modeling

[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.

[NeurIPS 2023] ImageReward: Learning and Evaluating Human Preferences for Text-to-image Generation

DDPO for finetuning diffusion models, implemented in PyTorch with LoRA support

Code release for H-GAP Humanoid Control with a Generalist Planner

[ICML 2024 Best Paper] Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution (https://arxiv.org/abs/2310.16834)

Code release for "Stochastic Optimal Control Matching"

A generative model for programmable protein design

A JAX-based simulator for autonomous driving research.

scikit-fmm is a Python extension module which implements the fast marching method.

Original reference implementation of "3D Gaussian Splatting for Real-Time Radiance Field Rendering"

code for "Riemannian Flow Matching on General Geometries".

Minimalist ML framework for Rust

Rust bindings for the C++ api of PyTorch.

High-performance quantum systems simulation with JAX (GPU-accelerated & differentiable solvers).

Mastering Diverse Domains through World Models

Normalizing flows in PyTorch

Official pytorch implementation for our ICLR 2023 paper "Latent State Marginalization as a Low-cost Approach for Improving Exploration".

Line-by-line profiling for Python

Auto-differentiable and hardware-accelerated force density method