Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

RDMA core userspace libraries and daemons

Bindings for RDMA ibverbs through rdma-core

SGLang is a fast serving framework for large language models and vision language models.

Tensors and Dynamic neural networks in Python with strong GPU acceleration

A machine learning compiler for GPUs, CPUs, and ML accelerators

LLM inference in C/C++

A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper GPUs, to provide better performance with lower memory utilization in bot…

A simple yet powerful tool to turn traditional container/OS images into unprivileged sandboxes.

Official NetHack Git Repository

A toolkit for developing and comparing reinforcement learning algorithms.

Multithreaded Python without the GIL

The Arcade Learning Environment (ALE) -- a platform for AI research.

libnop: C++ Native Object Protocols

Repository for the gRPC website and documentation

higher is a pytorch library allowing users to obtain higher order gradients over losses spanning training loops rather than individual training steps.

Convert nvprof profiles into about:tracing compatible JSON files

A TensorFlow implementation of Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures.