Doppel-8 · November 6, 2026
Doppel-8Paper availableCode and weightsDoppel-8 (doppel-emb8) · Non-commercial
Default checkpoint · Doppel-8 (doppel-emb8)
Participating organizationsCarnegie Mellon UniversityParticipating organizationsJohns Hopkins UniversityVersion scope and roles for these associations await verification.
Doppel is a cross-embodiment video world model from Johns Hopkins University, Carnegie Mellon University and Toyota Research Institute. Given an initial image, camera-aligned renders of a planned motion and its numerical trajectory, it predicts how the scene and objects will move.
Paper available
Author order: Original order
Shared render-action video world model and the eight-embodiment Doppel-8 experiment.
Publisher not provided · Code repository
Publication date unknown · Accessed: 2026-11-19
Publisher not provided · Official website
Publication date unknown · Accessed: 2026-11-19
Publisher not provided · Documentation
Publication date unknown · Accessed: 2026-11-19
Publisher not provided · Documentation
Publication date unknown · Accessed: 2026-11-19
Publisher not provided · Code repository
Publication date unknown · Accessed: 2026-11-19