MiniCPM-V 2.6: A GPT-4V Level MLLM for Single Image, Multi Image and Video on Your Phone

Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generation

Python based web automation tool. Powerful and elegant.

Maximize the Resolution Potential of Pre-trained Rectified Flow Transformers

Training open neural machine translation models

Fast inference engine for Transformer models

Scene Classification of 365 Scenes by fine-tuning Vision Transformer on Places365 Standard dataset

Expert Specialized Fine-Tuning

The data factory for next gen AI

A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep learning training and inference applications.

A linear estimator on top of clip to predict the aesthetic quality of pictures

Easily convert common crawl to a dataset of caption and document. Image/text Audio/text Video/text, ...

Workable training script for ControlNet tile

A very compact representation of a placeholder for an image.

Re-implementation of ControlNet with shape masks.

A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

C python bytecode disassembler and decompiler

Efficient Triton Kernels for LLM Training

Various AI scripts. Mostly Stable Diffusion stuff.

xDiT: A Scalable Inference Engine for Diffusion Transformers (DiTs) on multi-GPU Clusters

A microframework on top of PyTorch with first-class citizen APIs for foundation model adaptation

One-click Face Swapper and Restoration powered by insightface 🔥

[CVPR 2024] code release for "DiffusionLight: Light Probes for Free by Painting a Chrome Ball"

Open source implementation of AlphaFold3

TorchCFM: a Conditional Flow Matching library

It's React, but in Python

High-fidelity performance metrics for generative models in PyTorch