Meta Lingua: a lean, efficient, and easy-to-hack codebase to research LLMs.

💥 Fast State-of-the-Art Tokenizers optimized for Research and Production

Unsupervised text tokenizer for Neural Network-based text generation.

[NeurIPS 2024] Boosting the performance of consistency models with PCM!

Does VLM Classification Benefit from LLM Description Semantics? (AAAI 2025)

The official Pytorch implementation of “BAD: Bidirectional Auto-regressive Diffusion for Text-to-Motion Generation”

A framework for few-shot evaluation of language models.

Code for BLT research paper

[arXiv:2406.07548] Image and Video Tokenization with Binary Spherical Quantization

Code for https://arxiv.org/abs/2406.04329

official code for Diff-Instruct algorithm for one-step diffusion distillation

Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesis

[NeurIPS 2024] Official implementation of "Unlocking the Capabilities of Masked Generative Models for Image Synthesis via Self-Guidance"

Autonomous agents for everyone

A PyTorch library for implementing flow matching algorithms, featuring continuous and discrete flow matching implementations. It includes practical examples for both text and image modalities.

The official implementation of "[MASK] is All You Need"

(NeurIPS 2024 Oral 🔥) Improved Distribution Matching Distillation for Fast Image Synthesis

Bare-bones diffusion model code

Tiny AutoEncoder for Stable Diffusion

[ICLR2024] The official implementation of paper "VDT: General-purpose Video Diffusion Transformers via Mask Modeling", by Haoyu Lu, Guoxing Yang, Nanyi Fei, Yuqi Huo, Zhiwu Lu, Ping Luo, Mingyu Ding.

Codes for Visual Sketchpad: Sketching as a Visual Chain of Thought for Multimodal Language Models

DistillDIFT: Distillation of Diffusion Features for Semantic Correspondence (WACV 2025)

HunyuanVideo: A Systematic Framework For Large Video Generation Model

Official implementation of FIFO-Diffusion: Generating Infinite Videos from Text without Training (NeurIPS 2024)

20+ high-performance LLMs with recipes to pretrain, finetune and deploy at scale.

An open source implementation of the gameNgen paper

Code of Pyramidal Flow Matching for Efficient Video Generative Modeling