WorldBench WorldBench is a new benchmark designed to evaluate the physical understanding and prediction of modern world models and vision-language models. There are two components: Video based: This is a benchmark designed to evaluate video-to-video world foundation models such as Cosmos. It consists of videos 132 frames long of 425 simulated scenes with RGB, Normals, Depth, Flow, and Segmentations. Text based: This is a subset which adds text-based questions to 181 videos… See the full description on the dataset page: https://huggingface.co/datasets/worldbenchmark/IntuitivePhysics.
Use this model
Pull with QuantumShield
quantumshield pull worldbenchmark/IntuitivePhysics Verify integrity
quantumshield verify worldbenchmark/IntuitivePhysics pip install
pip install quantumshield && quantumshield pull worldbenchmark/IntuitivePhysics Unverified Model
This model has not been PQC-verified. File integrity cannot be guaranteed against quantum threats.
README.md
IntuitivePhysics
WorldBench WorldBench is a new benchmark designed to evaluate the physical understanding and prediction of modern world models and vision-language models. There are two components: Video based: This is a benchmark designed to evaluate video-to-video world foundation models such as Cosmos. It consists of videos 132 frames long of 425 simulated scenes with RGB, Normals, Depth, Flow, and Segmentations. Text based: This is a subset which adds text-based questions to 181 videos… See the full description on the dataset page: https://huggingface.co/datasets/worldbenchmark/IntuitivePhysics.
Intended Uses
This model is registered on the QuantaMrkt quantum-safe registry. This model has not yet been PQC-verified.
Quick Start
# Install the CLI pip install quantumshield # Pull the model quantumshield pull worldbenchmark/IntuitivePhysics # Verify file integrity quantumshield verify worldbenchmark/IntuitivePhysics
About
WorldBench WorldBench is a new benchmark designed to evaluate the physical understanding and prediction of modern world models and vision-language models. There are two components: Video based: This is a benchmark designed to evaluate video-to-video world foundation models such as Cosmos. It consists of videos 132 frames long of 425 simulated scenes with RGB, Normals, Depth, Flow, and Segmentations. Text based: This is a subset which adds text-based questions to 181 videos… See the full description on the dataset page: https://huggingface.co/datasets/worldbenchmark/IntuitivePhysics.
Get this model
Pull with QuantumShield
quantumshield pull worldbenchmark/IntuitivePhysics Verify signatures
quantumshield verify worldbenchmark/IntuitivePhysics