🤗A PyTorch-native Inference Engine with Hybrid Cache Acceleration and Parallelism for DiTs: Z-Image, FLUX2, Qwen-Image, etc.

Moshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec.

open-source multimodal large language model that can hear, talk while thinking. Featuring real-time end-to-end speech input and streaming audio output conversational capabilities.

Pytorch implementation of DoReMi, a method for optimizing the data mixture weights in language modeling datasets

Measuring Massive Multitask Language Understanding | ICLR 2021

Fast and flexible reference benchmarks

Grounded SAM: Marrying Grounding DINO with Segment Anything & Stable Diffusion & Recognize Anything - Automatically Detect , Segment and Generate Anything

Instruct-tune LLaMA on consumer hardware

LiBai(李白): A Toolbox for Large-Scale Distributed Parallel Training

A more efficient yolov5 with oneflow backend 🎉🎉🎉

OneFlow->ONNX

detrex is a research platform for DETR-based object detection, segmentation, pose estimation and other visual recognition tasks.

PyTorch implementation of MAE https//arxiv.org/abs/2111.06377

The implementation of “Gradient Harmonized Single-stage Detector” published on AAAI 2019.

You Only Look One-level Feature (YOLOF), CVPR2021, Detectron2

Two-stage CenterNet

Boosting your Web Services of Deep Learning Applications.

A PyTorch Toolbox for Face Recognition

[ICML2021] What Makes for End-to-End Object Detection

The official PyTorch implementation of paper BBN: Bilateral-Branch Network with Cumulative Learning for Long-Tailed Visual Recognition

ICCV19: Official code of Human Mesh Recovery from Monocular Images via a Skeleton-disentangled Representation

This repo is implemented based on detectron2 and centernet

A pytorch codebase for human parsing and vehicle parsing

Simplify your onnx model

Object detection achieving 44.3 mAP / 45 fps on COCO dataset

Monocular, One-stage, Regression of Multiple 3D People and their 3D positions & trajectories in camera & global coordinates. ROMP[ICCV21], BEV[CVPR22], TRACE[CVPR2023]

The best way to write secure and reliable applications. Write nothing; deploy nowhere.

support deepsort and bytetrack MOT(Multi-object tracking) using yolov5 with C++