Mixture-of-Experts (MoE) Language Model

Quantized Attention that achieves speedups of 2.1-3.1x and 2.7-5.1x compared to FlashAttention2 and xformers, respectively, without lossing end-to-end metrics across various models.

Together Mixture-Of-Agents (MoA) – 65.1% on AlpacaEval with OSS models

SSD-1B, an open-source text-to-image model, outperforming previous versions by being 50% smaller and 60% faster than SDXL.

[ICLR2025] Kolmogorov-Arnold Transformer

Lora dreambooth training with inpainting tuned SDXL model

🤗 Diffusers: State-of-the-art diffusion models for image and audio generation in PyTorch

Official implementation for "Targeted Cause Discovery with Data-Driven Learning"

[NAACL 2025] From redundancy to relevance: Enhancing explainability in multimodal large language models

Dermatology ddx dataset, Jax implementations of Monte Carlo conformal prediction, plausibility regions and statistical annotation aggregation from our recent work on uncertain ground truth (TMLR'23…

[NeurIPS 2024] Official Repository of The Mamba in the Llama: Distilling and Accelerating Hybrid Models

[ICLR 2025] Repository for Show-o, One Single Transformer to Unify Multimodal Understanding and Generation.

Eagle Family: Exploring Model Designs, Data Recipes and Training Strategies for Frontier-Class Multimodal LLMs

High-resolution models for human tasks.

🔍 우리가 읽을 논문을 찾아서, Cite.GG

This repository contains demos I made with the Transformers library by HuggingFace.

Understanding the Dependence of Perception Model Competency on Regions in an Image

Unofficial implementation of Palette: Image-to-Image Diffusion Models by Pytorch

KoLLaVA: Korean Large Language-and-Vision Assistant (feat.LLaVA)

Official Pytorch implementation of "Reference Guided Image Inpainting using Facial Attributes", BMVC 2021

This repository is the official repository of the GIM.

Scaling Diffusion Transformers with Mixture of Experts

Official PyTorch Implementation of MambaVision: A Hybrid Mamba-Transformer Vision Backbone

Mixture of A Million Experts

Official implementation of CVPR 2024 paper "Multi-criteria Token Fusion with One-step-ahead Attention for Efficient Vision Transformers".

AI Interviewer based on your CV

Official repository for "AM-RADIO: Reduce All Domains Into One"