Stars
Lightweight Image Video Action Generation Inference Framework
FireRed-Image-Edit is a powerful image editing foundation model achieving open-source state-of-the-art performance with precise instruction following, high-fidelity generation, superior identity co…
High-performance inference framework for large language models, focusing on efficiency, flexibility, and availability.
A lightweight design for computation-communication overlap.
SGLang is a high-performance serving framework for large language models and multimodal models.
A Flexible Framework for Experiencing Heterogeneous LLM Inference/Fine-tune Optimizations
HunyuanImage-3.0: A Powerful Native Multimodal Model for Image Generation
Distributed Compiler based on Triton for Parallel Systems
NVIDIA NVSHMEM is a parallel programming interface for NVIDIA GPUs based on OpenSHMEM. NVSHMEM can significantly reduce multi-process communication and coordination overheads by allowing programmer…
DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
Development repository for the Triton language and compiler
Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels
Fast and memory-efficient exact attention
FlashInfer: Kernel Library for LLM Serving
xDiT: A Scalable Inference Engine for Diffusion Transformers (DiTs) with Massive Parallelism
A flexible and efficient codebase for training visually-conditioned language models (VLMs)
Ongoing research training transformer models at scale
An open source implementation of CLIP.
CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image
LAVIS - A One-stop Library for Language-Vision Intelligence
InternLM-XComposer2.5-OmniLive: A Comprehensive Multimodal System for Long-term Streaming Video and Audio Interactions
A high-throughput and memory-efficient inference and serving engine for LLMs
Graph Neural Network Library for PyTorch
StyleGAN2 - Official TensorFlow Implementation
Code for "Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks"
A PyTorch Library for Meta-learning Research
FSL-Mate: A collection of resources for few-shot learning (FSL).