ZMIME

H3-Tiny-17B-FewStep-Ref2VA-v0.1 vs PAN 2

Compare documented capabilities and access. A dash means a comparable figure is unavailable.

Model specifications
SpecificationH3-Tiny-17B-FewStep-Ref2VA-v0.1H3-TinyPAN 2MBZUAI Institute of Foundation Models
TaskAudiovisual generationWorld simulation
Published capabilitiesGenerates video and synchronized sound from reference imagery using a pruned MiniMax H3 transformer.Predicts what happens after an action in a starting scene.
Model size17.08B transformer parameters—
AccessDownloadable transformer checkpoint for the H3 pipelinePublic weights, full training code and inference code
Result—Simulated video conditioned on a scene and action
Operating limitsSix steps by default—
SourcesModel weightsPublisher documentation
InputsText, ImagesText, Images
WeightsOpen weights · MiniMax H3 Community LicenseOpen weights · Apache-2.0
ReleasedNov 19, 2026Nov 10, 2026