Generate audio-reactive SCAIL pose sequences for character animation without requiring input video tracking.

A real-time streaming conversational video system that transforms text interactions into continuous, high-fidelity video responses using autoregressive diffusion.

official implementation of the paper "Delving into Latent Spectral Biasing of Video VAEs for Superior Diffusability".

Offical Implementation of SCAIL: Towards Studio-Grade Character Animation via In-Context Learning of 3D-Consistent Pose Representations

Official Implementations for Paper - HoloCine: Holistic Generation of Cinematic Multi-Shot Long Video Narratives

Kaleido: Open-sourced multi-subject reference video generation model, enabling controllable, high-fidelity video synthesis from multiple image references.

Official codebase for "Self Forcing: Bridging Training and Inference in Autoregressive Video Diffusion" (NeurIPS 2025 Spotlight)

Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

Video Prediction Policy: A Generalist Robot Policy with Predictive Visual Representations https://video-prediction-policy.github.io

(CVPR 2025) From Slow Bidirectional to Fast Autoregressive Video Diffusion Models

Concat-ID: Towards Universal Identity-Preserving Video Synthesis

[CVPR 2025] Hallo3: Highly Dynamic and Realistic Portrait Image Animation with Video Diffusion Transformer

[AAAI 2026] VisionReward: Fine-Grained Multi-Dimensional Human Preference Learning for Image and Video Generation

Official code for MotionBench (CVPR 2025)

[CVPR 2025 Highlight🔥] Identity-Preserving Text-to-Video Generation by Frequency Decomposition

Simple Controlnet module for CogvideoX model.

Helpful tools and examples for working with flex-attention

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

[CVPR 2025 Highlight] 3DTopia-XL: High-Quality 3D PBR Asset Generation via Primitive Diffusion

Next-Token Prediction is All You Need

Keyframe Interpolation with CogvideoX

Movie Gen Bench - two media generation evaluation benchmarks released with Meta Movie Gen

[CVPR'25]Tora: Trajectory-oriented Diffusion Transformer for Video Generation

Scalable and memory-optimized training of diffusion models

CogView4, CogView3-Plus and CogView3(ECCV 2024)

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

text and image to video generation: CogVideoX (2024) and CogVideo (ICLR 2023)