Adams Optimizer for Stable and Scalable Training

Efficient Triton Kernels for LLM Training

Hierarchical Reasoning Model Official Release

Monte Carlo tree search in JAX

Fused Adam-atan2 implementation

unofficial re-implementation of "Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets"

Github action to maximize the available disk space on Github runners

BFloat16 Fused Adam Operator for PyTorch

[AAAI'25 Oral] Are Expressive Models Truly Necessary for Offline RL?

Grok open release

Typed command line interfaces with argparse and pydantic

[For SM90 and cuBLAS] PyTorch bindings for CUTLASS grouped GEMM.

Official implementation for the paper: "Code Generation with AlphaCodium: From Prompt Engineering to Flow Engineering""

Pure C++ implementation of several models for real-time chatting on your computer (CPU & GPU)

Typed Argument Parsing with Pydantic

PyTorch bindings for CUTLASS grouped GEMM.

NVIDIA Linux open GPU kernel module source

ICLR 2022 Paper, SOTA Table Pre-training Model, TAPEX: Table Pre-training via Learning a Neural SQL Executor

The official repository for the paper "From Zero to Hero: Examining the Power of Symbolic Tasks in Instruction Tuning".

CVPR and NeurIPS poster examples and templates. May we have in-person poster session soon!

Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

OpenChat: Advancing Open-source Language Models with Imperfect Data

Code for paper Evolving Connectivity for Spiking Neural Networks

A multi-purpose LLM framework for RAG and data creation.

WireGuard Configuration Portal with LDAP connection