Stars
[NeurIPS 2025] Flow x RL. "ReinFlow: Fine-tuning Flow Policy with Online Reinforcement Learning". Support VLAs e.g., Pi0, Pi0.5, GR00TN1.5. Fully open-sourced.
A FlashAttention implementation for JAX with support for efficient document mask computation and context parallelism.
Xwin-LM: Powerful, Stable, and Reproducible LLM Alignment
Single-line inference of SOTA deep learning models
Library for reading and processing ML training data.
jax-triton contains integrations between JAX and OpenAI Triton
Infer.NET is a framework for running Bayesian inference in graphical models
YTsaurus is a scalable and fault-tolerant open-source big data platform.
Evals is a framework for evaluating LLMs and LLM systems, and an open-source registry of benchmarks.
Orbax provides common checkpointing and persistence utilities for JAX users
Nethack Learning Environment Wrapper for Language Interface
A toolkit for developing and comparing reinforcement learning algorithms.
Development repository for the Triton language and compiler
Flax is a neural network library for JAX that is designed for flexibility.
High-quality implementations of standard and SOTA methods on a variety of tasks.
A concise but complete full-attention transformer with a set of promising experimental features from various papers
Pyrallis is a framework for structured configuration parsing from both cmd and files. Simply define your desired configuration structure as a dataclass and let pyrallis do the rest!
State-of-the-Art Deep Learning scripts organized by models - easy to train and deploy with reproducible accuracy and performance on enterprise-grade infrastructure.
MiniHack the Planet: A Sandbox for Open-Ended Reinforcement Learning Research
Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more