gzhu06

mlp

The Multilayer Perceptron Language Model

rectified-flow-pytorch

Implementation of rectified flow and some of its followup research / improvements in Pytorch

Kolors

Kolors Team

rfpp

The codebase of our paper "Improving the Training of Rectified Flows"

disco-diffdock

Code for the paper DisCo-Diff: Enhancing Continuous Diffusion Models with Discrete Latents, ICML 2024

min-fsdp

E-LatentLPIPS

Unofficial Implementation of E-LatentLPIPS(Ensembled-LatentLPIPS) of Diffusion2GAN

LLM101n

LLM101n: Let's build a Storyteller

titok-pytorch

Implementation of TiTok, proposed by Bytedance in "An Image is Worth 32 Tokens for Reconstruction and Generation"

build-nanogpt

Video+code lecture on building nanoGPT from scratch

Omost

Your image is almost there!

schedule_free

Schedule-Free Optimization in PyTorch

vit-pytorch

Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch

conditional-flow-matching

fms-fsdp

🚀 Efficiently (pre)training foundation models with native PyTorch features, including FSDP for training and SDPA implementation of Flash attention v2.

Langevin_Dynamics_Sampling_2D

Langevin_Dynamics_Sampling_2D

Lumina-T2X

Lumina-T2X is a unified framework for Text to Any Modality Generation

cookbook

Deep learning for dummies. All the practical details and useful utilities that go into working with real models.

minRF

Minimal implementation of scalable rectified flow transformers, based on SD3's approach

imageinwords

Data release for the ImageInWords (IIW) paper.

VAR

[GPT beats diffusion🔥] [scaling laws in visual generation📈] Official impl. of "Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction". An *ultra-simple, user-friendly yet state-of-the-art* codebase for autoregressive image generation!

llama3

The official Meta Llama 3 GitHub site

ollama

Get up and running with Llama 3, Mistral, Gemma 2, and other large language models.

d3pm

Minimal Implementation of a D3PM in pytorch

DOVER

[ICCV 2023, Official Code] for paper "Exploring Video Quality Assessment on User Generated Contents from Aesthetic and Technical Perspectives". Official Weights and Demos provided.

llm.c

LLM training in simple, raw C/CUDA

Sophia

The official implementation of “Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-training”

xformers

Hackable and optimized Transformers building blocks, supporting a composable construction.

FunCodec

FunCodec is a research-oriented toolkit for audio quantization and downstream applications, such as text-to-speech synthesis, music generation et.al.

Gull