Unified Efficient Fine-Tuning of 100+ LLMs (ACL 2024)

Bag of Tricks and A Strong Baseline for Deep Person Re-identification

DINO-X: The World's Top-Performing Vision Model for Open-World Object Detection and Understanding

A Versatile Video-LLM for Long and Short Video Understanding with Superior Temporal Localization Ability

Augmenting Multimodal LLMs with Self-Reflective Tokens for Knowledge-based Visual Question Answering

Project for "HyperSeg: Towards Universal Visual Segmentation with Large Language Model".

Fast and memory-efficient exact attention

[ECCV 2024] Decomposition Betters Tracking Everything Everywhere

Offical code of "QKFormer: Hierarchical Spiking Transformer using Q-K Attention" (NeurIPS 2024，Spotlight 3%)

Linux 0.01源码及注释

Historically first (actually 0.01) release of linux kernel made by Linus Torvalds

xLSTM as Generic Vision Backbone

Continuation of Clash Verge - A Clash Meta GUI based on Tauri (Windows, MacOS, Linux)

Python 3.8+ toolbox for submitting jobs to Slurm

The repository provides code for running inference with the Meta Segment Anything Model 2 (SAM 2), links for downloading the trained model checkpoints, and example notebooks that show how to use th…

Tutel MoE: An Optimized Mixture-of-Experts Implementation

This is a code repository of paper "Edge-cloud Collaborative Learning with Federated andCentralized Features"

edge-cloud learning framework and experiments

A PyTorch Implementation for experiements in paper: Neurosurgeon: Collaborative Intelligence Between the Cloud and Mobile Edge.

A MoE impl for PyTorch, [ATC'23] SmartMoE

Ongoing research training transformer models at scale

A fast MoE impl for PyTorch

[ECCV2024] Video Foundation Models & Data for Multimodal Understanding

OpenMMLab's Next Generation Video Understanding Toolbox and Benchmark

[NeurIPS 2022 Spotlight] VideoMAE for Action Detection

This repository is an official implementation of the paper "LW-DETR: A Transformer Replacement to YOLO for Real-Time Detection".

WEB VIDEO PLATFORM是一个基于GB28181-2016标准实现的网络视频平台，支持NAT穿透，支持海康、大华、宇视等品牌的IPC、NVR、DVR接入。支持国标级联，支持rtsp/rtmp等视频流转发到国标平台，支持rtsp/rtmp等推流转发到国标平台。