PyTorch code and models for V-JEPA self-supervised learning from video.

Official codebase for I-JEPA, the Image-based Joint-Embedding Predictive Architecture. First outlined in the CVPR paper, "Self-supervised learning from images with a joint-embedding predictive architecture."

An open source library designed to provide community examples of Joint Embedding Predictive Architectures (JEPAs). It contains code and examples for learning representations from images, video, and action-conditioned video, as well as planning using JEPA-based models.

Open source repo for Locate 3D Model, 3D-JEPA and Locate 3D Dataset

Reliable, minimal and scalable library for evaluating and conducting world model research

Implementation of I-JEPA from "Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture"

EchoJEPA: A Latent Predictive Foundation Model for Echocardiography

This repo contains the code for the paper "Intuitive physics understanding emerges fromself-supervised pretraining on natural videos"

No description available

Code, data and weights for the paper **What drives success in physical planning with Joint-Embedding Predictive World Models?**

Official codebase for "Brain-JEPA: Brain Dynamics Foundation Model with Gradient Positioning and Spatiotemporal Masking" (NeurIPS 2024, Spotlight).

VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Model

No description available

Repo of "Drive-JEPA: Video JEPA Meets Multimodal Trajectory Distillation for End-to-End Driving"

Self-Supervised Learning for SAR ATR with a Joint-Embedding Predictive Architecture

VL-JEPA (Vision-Language Joint Embedding Predictive Architecture) in MLX

Official Code for Rectified LpJEPA: Joint-Embedding Predictive Architectures with Sparse and Maximum-Entropy Representations

Joint Embedding Predictive Architecture for Musical Stem Compatibility Estimation

Experiments in Joint Embedding Predictive Architectures (JEPAs).

Audio-JEPA is an adaptation of the Joint-Embedding Predictive Architecture (JEPA) for self-supervised audio representation learning. Built upon the I-JEPA paradigm, it uses a Vision Transformer (ViT) backbone to predict latent representations of masked spectrogram patches.

This is the official repository for the paper: JEPA-T: Joint-Embedding Predictive Architecture with Text Fusion for Image Generation

Official code for the paper Graph-level Representation Learning with Joint-Embedding Predictive Architectures published in Transactions on Machine Learning Research (TMLR)

[WACV 2025] Point-JEPA: A Joint Embedding Predictive Architecture for Self-Supervised Learning on Point Cloud

[AAAI 2026] AD-L-JEPA: Self-Supervised Representation Learning with Joint Embedding Predictive Architecture for Automotive LiDAR Object Detection

No description available

👆PyTorch Implementation of JEDi Metric described in "Beyond FVD: Enhanced Evaluation Metrics for Video Generation Quality"

Official implementation of our paper "CNN-JEPA: Self-Supervised Pretraining Convolutional Neural Networks Using Joint Embedding Predictive Architecture"

No description available

Implementation of "TS-JEPA"

A Predictive World Model of the Transcriptome