Glint is a Rust framework designed for creating stateful, graph-based AI systems, enabling efficient multi-step workflows. With features like LLM integration and a graph-based architecture, Glint helps developers build powerful AI solutions with ease. 🐙✨

The Shamrock Framework, an open-source, multi-GPU hydrodynamics framework for astrophysics. Scales seamlessly from laptops to exascale supercomputers, supporting SPH, AMR, and more.

👤 Implement face recognition using TensorFlow, featuring advanced techniques for accurate identification and clustering of faces in images.

QUDA is a library for performing calculations in lattice QCD on GPUs.

Distributed tensors and Machine Learning framework with GPU and MPI acceleration in Python

An versatile Open Core agentic specialist scaffold for building agentic systems with LangGraph.

TransCorpus is a scalable toolkit for large-scale, parallel translation and preprocessing of text corpora, built for language model pretraining and research.

Efficient Distributed GPU Programming for Exascale, an SC/ISC Tutorial

XReflection is a neat toolbox tailored for single-image reflection removal(SIRR). We offer state-of-the-art SIRR solutions for training and inference, with a high-performance data pipeline, multi-GPU/TPU/NPU support, and more!

Almost trivial distributed parallelization of stencil-based GPU and CPU applications on a regular staggered grid

A dual-GPU DEM solver with complex grain geometry support

Designed for open-weights LLMs to test capabilities using BFCL tests.

Clara AI System - Machine learning with continuous training, multi-GPU support and QLoRA fine-tuning

🌌 htop/btop is yesterday. 🦀 Rust-Powered | 🌌 Cyberpunk Aesthetics | 🤖 Predictive Analytics | 🎮 130+ GPU Models (NVIDIA/AMD/Intel) | 📊 Real-Time Alerts | ⚡ Zero Dependencies | For developers who refuse boring dashboards.

GPU-accelerated linear solvers based on the conjugate gradient (CG) method, supporting NVIDIA and AMD GPUs with GPU-aware MPI, NCCL, RCCL or NVSHMEM

TorchDR - PyTorch Dimensionality Reduction

Repository of Computer Vision projects based on CNNs, Vision Transformers, and YOLO11, implemented with TensorFlow, PyTorch, Hugging Face, and Ultralytics.

POT3D: High Performance Potential Field Solver

Real-time N-Body Algorithm Accelerated With CUDA and FFT (for up to 1 billion particles with multi GPU)