Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

🔥🔥 Official Repo of UMO: Scaling Multi-Identity Consistency for Image Customization via Matching Reward

Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference

HunyuanImage-2.1: An Efficient Diffusion Model for High-Resolution (2K) Text-to-Image Generation​

Your financial investment assistant

A parallelism VAE avoids OOM for high resolution image generation

xDiT: A Scalable Inference Engine for Diffusion Transformers (DiTs) with Massive Parallelism

A unified inference and post-training framework for accelerated video generation.

An official implementation of DanceGRPO: Unleashing GRPO on Visual Generation

Official implementation of Pref-GRPO: Pairwise Preference Reward-based GRPO for Stable Text-to-Image Reinforcement Learning

🔥🔥 Open-sourced unified customization model

Home Made Diffusion Models

[Official] Voost: A Unified and Scalable Diffusion Transformer for Bidirectional Virtual Try-On and Try-Off

Reference PyTorch implementation and models for DINOv3

Embedding Atlas is a tool that provides interactive visualizations for large embeddings. It allows you to visualize, cross-filter, and search embeddings and metadata.

[NeurIPS 2025] An official implementation of Flow-GRPO: Training Flow Matching Models via Online RL

A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit floating point (FP8) precision on Hopper, Ada and Blackwell GPUs, to provide better performance with lower memory…

Qwen-Image is a powerful image generation foundation model capable of complex text rendering and precise image editing.

Official repository of "Beyond Fixed: Training-Free Variable-Length Denoising for Diffusion Large Language Models"

Wan: Open and Advanced Large-Scale Video Generative Models

[NeurIPS 2025] Official implementation of "XVerse: Consistent Multi-Subject Control of Identity and Semantic Attributes via DiT Modulation".

Ming - facilitating advanced multimodal understanding and generation capabilities built upon the Ling LLM.

SKT A.X LLM 4.0

The official repo for “Dolphin: Document Image Parsing via Heterogeneous Anchor Prompting”, ACL, 2025.

Official Implementation of OCR-free Document Understanding Transformer (Donut) and Synthetic Document Generator (SynthDoG), ECCV 2022

MiniMax-M1, the world's first open-weight, large-scale hybrid-attention reasoning model.

MiniCPM4 & MiniCPM4.1: Ultra-Efficient LLMs on End Devices, achieving 3+ generation speedup on reasoning tasks