allenai/molmo-motion

Official code of Motus: A Unified Latent Action World Model

[NeurIPS 2023] MotionGPT: Human Motion as a Foreign Language, a unified motion-language generation model using LLMs

Code for the Molmo2 Vision-Language Model

Official implementation of "Learning How the World Evolves: Extrapolative Video World Models via Latent Dynamics Reasoning"

Official implementation of PhiZero: A World Model Built Around Physical Language. Code and checkpoints coming soon.

code for the paper Predicting Point Tracks from Internet Videos enables Diverse Zero-Shot Manipulation

TempoVista is the official implementation of Dyn-3D, a benchmark and training pipeline for 3D spatial reasoning in vision-language models. The release contains frame-selection tools, SFT and Kinematic-GSPO training code, benchmark evaluation scripts, Dyn-3D annotations, and checkpoints for Qwen3-VL and InternVL3.5.

Official implementation of Video Prediction Policy 2: Predict Better, Act Better

[NeurIPS 2024 Spotlight] Motion Forecasting in Continuous Driving