← Local AI Release Board

nvidia/PixelUMM

main boardmodel

PixelUMM is an NVIDIA-developed, encoder-free unified multimodal model for joint understanding and generation across text, images, and video directly in pixel space. License: nvidia-one-way-noncommercial-license.

Org
nvidia
Released
2026-10-01
Status
main board
License
nvidia-one-way-noncommercial-license (restricted)
Task
any-to-any · omni
Signal
89 HF likes · 0 downloads
Weights
no weight files listed

Links

First published here 2026-10-01 · updated 2026-10-01 · markdown · JSON