#
Starred repositories
《开源大模型食用指南》针对中国宝宝量身打造的基于Linux环境快速微调（全参数/Lora）、部署国内外开源大模型（LLM）/多模态大模型（MLLM）教程
Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)
Dead simple FLUX LoRA training UI with LOW VRAM support
Repository of the paper "AnyUp: Universal Feature Upsampling".
[CVPR '24] Official Implementation of Flow-Guided Online Stereo Rectification for Wide Baseline Stereo
This project implements knowledge distillation from DINOv2 (Vision Transformer) to convolutional networks, enabling efficient visual representation learning with reduced computational requirements.
Pytorch implementation of various Knowledge Distillation (KD) methods.
A PyTorch implementation for exploring deep and shallow knowledge distillation (KD) experiments with flexibility
[NeurIPS 2025] Pixel-Perfect Depth
The official project website of "ScaleKD: Strong Vision Transformers Could Be Excellent Teachers" (ScaleKD for short, accepted to NeurIPS 2024).
[DEIMv2] Real Time Object Detection Meets DINOv3
Hunyuan 3D Part Segmentation and Generation Pipeline
The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (V…
The AI developer platform. Use Weights & Biases to train and fine-tune models, and manage models from experimentation to production.
Sim-to-real and CDM inference code for ManipAsInSim project.
Source code for ICCV 2025 paper "FlowSeek: Optical Flow Made Easier with Depth Foundation Models and Motion Bases"
[CVPR 2025 Highlight] Video Depth Anything: Consistent Depth Estimation for Super-Long Videos
[CVPR 2025 Highlight] GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control
ComfyUI native implementation of IC-Light
Infinite Photorealistic Worlds using Procedural Generation
Official implementation of ICCV2023 VideoFlow: Exploiting Temporal Cues for Multi-frame Optical Flow Estimation
Learning to Estimate Hidden Motions with Global Motion Aggregation (ICCV 2021)
The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface.