[ICLR 2026] Thinking with Camera: A Unified Multimodal Model for Camera-Centric Understanding and Generation

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

StreamingVLM: Real-Time Understanding for Infinite Video Streams

[NeurIPS 2023] ImageReward: Learning and Evaluating Human Preferences for Text-to-image Generation

SigLIP-based Aesthetic Score Predictor

Carbon Language's main repository: documents, design, implementation, and related tools. (NOTE: Carbon Language is experimental; see README)

Turn any computer or edge device into a command center for your computer vision projects.

[ICLR 2026] GenCompositor: Generative Video Compositing with Diffusion Transformer

[ICLR 2026] Streamlining Cartoon Production with Generative Post-Keyframing

Lumina-DiMOO - An Open-Sourced Multi-Modal Large Diffusion Language Model

[ICLR 2026] Official Repo for Rolling Forcing: Autoregressive Long Video Diffusion in Real Time

[ICML2025, NeurIPS2025 Spotlight] Sparse VideoGen 1 & 2: Accelerating Video Diffusion Transformers with Sparse Attention

Online resources for Python Crash Course, 3rd edition, from No Starch Press.

DC-VideoGen: Efficient Video Generation with Deep Compression Video Autoencoder

FlashInfer: Kernel Library for LLM Serving

DC-Gen: Post-Training Diffusion Acceleration with Deeply Compressed Latent Space

⚡️SwanLab - an open-source, modern-design AI training tracking and visualization tool. Supports Cloud / Self-hosted use. Integrated with PyTorch / Transformers / verl / LLaMA Factory / ms-swift / U…

LiteRT, successor to TensorFlow Lite. is Google's On-device framework for high-performance ML & GenAI deployment on edge platforms, via efficient conversion, runtime, and optimization

Nano vLLM

Generate a timeline of your day, automatically

[ICCV 2025] Official pytorch implementation of "FramePainter: Endowing Interactive Image Editing with Video Diffusion Priors"

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

Reference PyTorch implementation and models for DINOv3

🚀 「大模型」1小时从0训练26M参数的视觉多模态VLM！🌏 Train a 26M-parameter VLM from scratch in just 1 hours!

Diffusion model(SD,Flux,Wan,Qwen Image,Z-Image,...) inference in pure C/C++

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

Generative Omnimatte (CVPR 2025)