PyTorch template for Deep Learning projects with support for scalable multi-GPU and multi-node training.

A high-throughput and memory-efficient inference and serving engine for LLMs

Optimized primitives for collective multi-GPU communication

DeepEP: an efficient expert-parallel communication library

[DEPRECATED] Moved to ROCm/rocm-systems repo

UCCL is an efficient communication library for GPUs, covering collectives, P2P (e.g., KV cache transfer, RL weight transfer), and EP (e.g., GPU-driven)

FlashInfer: Kernel Library for LLM Serving

A framework for GPU-friendly implementations of tree algorithms -- such as kNN and FoF -- in jax (with a CUDA backend)

Open-source simulator for autonomous driving research.

CUDA-L2: Surpassing cuBLAS Performance for Matrix Multiplication through Reinforcement Learning

A rclcpp-compatible true zero-copy IPC middleware that supports all ROS message types, including message structs already generated by rosidl.

HSE course about CUDA programming

SGLang is a high-performance serving framework for large language models and multimodal models.

Official Kubernetes Operator for ClickHouse®

A profiling tool for agentic tools like Claude Code and Codex

Домашние задания для курса "Вычисления на видеокартах" в CS Space

The auto solution checking system for YSDA; server, storing grades and managing deadlines

Such tasks many wow

The First Real-Time Search Analytics Database

The best ChatGPT that $100 can buy.

A minimal PyTorch re-implementation of the OpenAI GPT (Generative Pretrained Transformer) training

Extension for Greenplum to read Apache Iceberg

Apache Iceberg C++ library

DuckDB extension for reading data stored in the Apache GraphAr format.

Dig into ClickHouse with TUI interface.

A Go implementation of EIP-4361 Sign In With Ethereum verification

Error-bounded Lossy Data Compressor (for floating-point/integer datasets)