A curated list of research papers, resources, and advancements on Diffusion Cache and related efficient diffusion model acceleration techniques.

Krea Realtime 14B. An open-source realtime AI video model.

DiffusionNFT: Online Diffusion Reinforcement with Forward Process

Trace Anything: Representing Any Video in 4D via Trajectory Fields

Follow-Your-Preference: Towards Preference-Aligned Image Inpainting

Lyra: Generative 3D Scene Reconstruction via Video Diffusion Model Self-Distillation

The official Pytorch code for paper "ContextFlow: Training-Free Video Object Editing via Adaptive Context Enrichment"

MapAnything: Universal Feed-Forward Metric 3D Reconstruction

Code for ICCVW paper, UniPaint: Unified Space-time Video Inpainting via Mixture-of-Experts

4DNeX: Feed-Forward 4D Generative Modeling Made Easy

Implementation of "S^2-Guidance: Stochastic Self Guidance for Training-Free Enhancement of Diffusion Models"

[CVPR 2024 Highlight] OpenESS: Event-Based Semantic Scene Understanding with Open Vocabularies

Cosmos-Transfer1-DiffusionRenderer: High-quality video de-lighting and re-lighting based on Cosmos video diffusion framework

Implementation of CamTrol: Training-free Camera Control for Video Generation

📹 A more flexible framework that can generate videos at any resolution and creates videos from images.

[ICCV'25] Official implementation of "Reangle-A-Video: 4D Video Generation as Video-to-Video Translation"

Official Implementation of "Instance Segmentation of Scene Sketches Using Natural Image Priors" (SIGGRAPH 2025)

ViPE: Video Pose Engine for Geometric 3D Perception

Uni3C: Unifying Precisely 3D-Enhanced Camera and Human Motion Controls for Video Generation [Siggraph Asian 2025]

[ArXiv 2025] Follow-Your-Shape: This repo is the official implementation of "Follow-Your-Shape: Shape-Aware Image Editing via Trajectory-Guided Region Control"

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

Code of π^3: Permutation-Equivariant Visual Geometry Learning

Wan: Open and Advanced Large-Scale Video Generative Models

Long-RL: Scaling RL to Long Sequences (NeurIPS 2025)

[ICCV 2025] SpatialTrackerV2: 3D Point Tracking Made Easy

Cosmos-Predict2 is a collection of general-purpose world foundation models for Physical AI that can be fine-tuned into customized world models for downstream applications.

[NeurIPS' 2025] JarvisArt: Liberating Human Artistic Creativity via an Intelligent Photo Retouching Agent

Calligrapher: Freestyle Text Image Customization