Enjoy the magic of Diffusion models!

Timestep Embedding Tells: It's Time to Cache for Video Diffusion Model

Official repository for our work on micro-budget training of large-scale diffusion models.

[ICLR 2025] FasterCache: Training-Free Video Diffusion Model Acceleration with High Quality

Official Pytorch Implementation for "VidToMe: Video Token Merging for Zero-Shot Video Editing" (CVPR 2024)

Official repository of In-Context LoRA for Diffusion Transformers

[ICLR 2025] OpenVid-1M: A Large-Scale High-Quality Dataset for Text-to-video Generation

Neighborhood Attention Transformer, arxiv 2022 / CVPR 2023. Dilated Neighborhood Attention Transformer, arxiv 2022

FastVideo is a lightweight framework for accelerating large video diffusion models.

SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformer

Inference-time scaling of diffusion-based image and video generation models.

[CVPR2025] A benchmark for evaluating video generative models in generating short stories

A list of works on evaluation of visual generation models, including evaluation metrics, models, and systems

Explore the Multimodal “Aha Moment” on 2B Model

Awesome Reasoning LLM Tutorial/Survey/Guide

s1: Simple test-time scaling

A general framework for inference-time scaling and steering of diffusion models with arbitrary rewards.

Diffusion-Sharpening: Fine-tuning Diffusion Models with Denoising Trajectory Sharpening

Inference-Time Alignment in Protein Diffusion Models

[NeurIPS 2024] SimPO: Simple Preference Optimization with a Reference-Free Reward

Wan: Open and Advanced Large-Scale Video Generative Models

R1-onevision, a visual language model capable of deep CoT reasoning.

[CVPR 2024] | LAMP: Learn a Motion Pattern for Few-Shot Based Video Generation

SkyReels V1: The first and most advanced open-source human-centric video foundation model

FlashVideo: Flowing Fidelity to Detail for Efficient High-Resolution Video Generation

Solve Visual Understanding with Reinforced VLMs