A generative world for general-purpose robotics & embodied AI learning.

Showing how the SDXL latent space corrections work

A family of compressed models obtained via pruning and knowledge distillation

Geometry Transfer for Stylizing Radiance Fields

Kolmogorov Arnold Networks

DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models

Implementation of Agent Attention in Pytorch

Diffusion Reinforcement Learning Library

Diffusion Reinforcement Learning Library

CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image

Collection and Implementation of Mobile-based Vision Transformer in Pytorch

Implementation of a memory efficient multi-head attention as proposed in the paper, "Self-attention Does Not Need O(n²) Memory"

Fast and memory-efficient exact attention

Visual Taste Approximator (VTA) is a very simple tool that helps anyone create an automatic replica of themselves that can approximate their own personal visual taste

A small CLI app to scrap high-quality movie snapshots from various websites.

Implementation of Muse: Text-to-Image Generation via Masked Generative Transformers, in Pytorch

An official implementation of MobileStyleGAN in PyTorch

[SIGGRAPH'22] StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasets

Easily compute clip embeddings and build a clip retrieval system with them

Code Release for MViTv2 on Image Recognition.

HRViT ("Multi-Scale High-Resolution Vision Transformer for Semantic Segmentation"), CVPR 2022.

Karras et al. (2022) diffusion models for PyTorch

CLOOB Conditioned Latent Diffusion training and inference code

A JAX nn library

Recent Transformer-based CV and related works.

RUDOLPH: One Hyper-Tasking Transformer can be creative as DALL-E and GPT-3 and smart as CLIP

Cellular automata based on randomly generated "color transition rules". Written in Rust as a learning project.