ZMIME

LongCat-Video-Avatar 1.6 vs PAN 2

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

Model specifications
SpecificationLongCat-Video-Avatar 1.6Meituan (LongCat)PAN 2MBZUAI Institute of Foundation Models
TaskAudio-driven character animationWorld simulation
Published capabilitiesAnimates speakers from audio and optional reference imagery, including multi-person scenes and continued shots.Predicts what happens after an action in a starting scene.
AccessDownloadable checkpoint and reference inference implementationPublic weights, full training code and inference code
Result—Simulated video conditioned on a scene and action
SourcesModel weights · Additional documentationPublisher documentation
InputsText, Images, AudioText, Images
WeightsOpen weights · MITOpen weights · Apache-2.0
ReleasedNov 15, 2026Nov 10, 2026