Showing 1-18 of 18 projects
Colossal-AI optimizes large AI model training and inference with distributed computing and GPU acceleration.
A flexible framework for optimizing heterogeneous LLM inference and fine-tuning workflows.
A high-performance task-parallel programming system for C++ developers building concurrent and heterogeneous applications.
Dolphin is a document image parsing library that uses heterogeneous anchor prompting for OCR and layout analysis.
A C++ interface for portability across different heterogeneous computing platforms like CUDA and HIP.
Project-HAMi/HAMi is a CNCF project for GPU virtualization middleware to enable GPU acceleration in Kubernetes clusters.
Specification for streaming massive heterogeneous 3D geospatial datasets across the web.
A community-driven C++ compiler for heterogeneous programming models like SYCL, C++ parallelism, and CUDA/HIP.
A trusted, open-source network access control (NAC) solution with powerful BYOD management and 802.1X support.
Official repository for a paper on a large vision-language model for medical applications
An efficient and flexible graph learning framework based on PaddlePaddle, with support for heterogeneous graphs and meta-path learning.
Parallel graph management and execution in heterogeneous computing
TornadoVM is a heterogeneous programming framework that enables developers to leverage GPUs and other accelerators for improved performance in Java applications.
A heterogeneous graph neural network library for advanced graph representation learning.
OpenCL integration for Python, with advanced features for parallel computing and GPU acceleration.
DataLink is a real-time and offline data exchange platform that supports synchronization between heterogeneous data sources.
pocl is a portable open-source implementation of the OpenCL standard for parallel programming on heterogeneous systems.
A high-performance, GPU-accelerated multimedia/video processing framework for transcoding, AI inference, and more.
Get weekly updates on trending AI coding tools and projects.