HPC-AI Tech ORG

Making large AI models cheaper, faster and more accessible

Open-Sora: Democratizing Efficient Video Production for All

Large-scale model inference.

Optimizing AlphaFold Training and Inference on GPU Clusters

Efficient AI Inference & Serving

Examples of training models with hybrid parallelism using ColossalAI

Scalable PaLM implementation of PyTorch

A Python library transfers PyTorch tensors between CPU and NVMe

A memory efficient DLRM training solution using ColossalAI

Sky Computing: Accelerating Geo-distributed Computing in Federated Learning

Performance benchmarking with ColossalAI

A collection of models built with ColossalAI

Documentation for Colossal-AI

A collection of dockerfiles for various tasks

No description.

Elixir: Train a Large Language Model on a Small GPU Cluster

Storing publicly available assets such as images, animations and texts

CLI for ColossalAI Platform

🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

GPT Demo with hybrid distributed training

No description.

HPC-AI TECH 's Fine-tuning SDK

No description.

Train mmdetection models with ColossalAI.

Democratizing AlphaFold3: an PyTorch reimplementation to accelerate protein structure prediction

Pytorch domain library for recommendation systems

A modular graph-based Retrieval-Augmented Generation (RAG) system

No description.

This repository contains Huawei Ascend CANN files

Documentation for our cloud platform

TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and support state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT LLM also contains components to create Python and C++ runtimes that orchestrate the inference execution in performant way.

A unified library of state-of-the-art model optimization techniques like quantization, pruning, distillation, speculative decoding, etc. It compresses deep learning models for downstream deployment frameworks like TensorRT-LLM or TensorRT to optimize inference speed.

Selects tests affected by changed files. Executes the right tests first. Continuous test runner when used with pytest-watch.