Native Multimodal Models are World Learners

The best ChatGPT that $100 can buy.

RF-DETR is a real-time object detection and segmentation model architecture developed by Roboflow, SOTA on COCO and designed for fine-tuning.

Cosmos-Predict2.5, the latest version of the Cosmos World Foundation Models (WFMs) family, specialized for simulating and predicting the future state of the world in the form of video.

[CVPR 2025] Code for Segment Any Motion in Videos

[ECCV2024 - Oral, Best Paper Award Candidate] SEA-RAFT: Simple, Efficient, Accurate RAFT for Optical Flow

GLM-4 series: Open Multilingual Multimodal Chat LMs | 开源多语言多模态对话模型

Reference PyTorch implementation and models for DINOv3

Wan: Open and Advanced Large-Scale Video Generative Models

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

MAGI-1: Autoregressive Video Generation at Scale

A beautiful, simple, clean, and responsive Jekyll theme for academics

An open-source AI agent that brings the power of Gemini directly into your terminal.

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.

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

[ICLR 2025][arXiv:2406.07548] Image and Video Tokenization with Binary Spherical Quantization

Official PyTorch Implementation of "Diffusion Autoencoders are Scalable Image Tokenizers"

🦜🔗 The platform for reliable agents.

Wan: Open and Advanced Large-Scale Video Generative Models

Cosmos-Reason1 models understand the physical common sense and generate appropriate embodied decisions in natural language through long chain-of-thought reasoning processes.

Cosmos-Predict1 is a collection of general-purpose world foundation models for Physical AI that can be fine-tuned into customized world models for downstream applications.

Cosmos-Transfer1 is a world-to-world transfer model designed to bridge the perceptual divide between simulated and real-world environments.

[NeurIPS 2024]OmniTokenizer: one model and one weight for image-video joint tokenization.

Enjoy the magic of Diffusion models!

Perceptual video quality assessment based on multi-method fusion.

This repo contains the code for 1D tokenizer and generator

ElasticTok: Adaptive Tokenization for Image and Video