World model PAN2 trends on Hugging Face: image-text-to-video generation

IFM · hf · 2026-11-16

PAN2 from IFM is trending on Hugging Face as an image-text-to-video model framed as a world model for world simulation. Built on a diffusion-transformer architecture, it supports both text-to-video and image/text-to-video generation, is available via diffusers with safetensors weights, and links to an arXiv paper (2507.05169).

Original post →

More from Multimodal

Multimodal channel →