Train transformer language models with reinforcement learning.

CUDA Templates for Linear Algebra Subroutines

PyTorch library for solving imaging inverse problems using deep learning

Boltzmann Generators and Normalizing Flows in PyTorch

PyTorch implementation of normalizing flow models

POT : Python Optimal Transport

A simple implimentation of Bayesian Flow Networks (BFN)

TorchCFM: a Conditional Flow Matching library

Reference implementation for Deep Unsupervised Learning using Nonequilibrium Thermodynamics

Bridging deep learning and logical reasoning using a differentiable satisfiability solver.

A PyTorch native library for large model training

[EMNLP'23, ACL'24] To speed up LLMs' inference and enhance LLM's perceive of key information, compress the prompt and KV-Cache, which achieves up to 20x compression with minimal performance loss.

[ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding

awesome synthetic (text) datasets

Official PyTorch implementation of Learning to (Learn at Test Time): RNNs with Expressive Hidden States

Tools for merging pretrained large language models.

[ICLR 2025] OpenVid-1M: A Large-Scale High-Quality Dataset for Text-to-video Generation

A framework for few-shot evaluation of language models.

Efficient Triton Kernels for LLM Training

A throughput-oriented high-performance serving framework for LLMs

[ICLR 2024 Spotlight] Official implementation of ScaleCrafter for higher-resolution visual generation at inference time.

AIOS: AI Agent Operating System

[ICCV 2023] StableVideo: Text-driven Consistency-aware Diffusion Video Editing

Official inference repo for FLUX.1 models

Accelerated First Order Parallel Associative Scan

The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM 2), links for downloading the trained model checkpoints, and example notebooks that show how to use th…

The Memory layer for AI Agents