hugpadhugpad

xiaomi-research / SpatioLM-Action-VLA0

Vision2.3B241Updated Aug 4, 2026
Hugging Face

Downloads

24

Likes

1

Last updated

Aug 4, 2026

License

apache-2.0

This is the official SpatioLM Action checkpoint based on SenseNova-SI 1.1 / InternVL3 (2B). It is intended for embodied manipulation with discrete VLA actions.

Tasks & library

  • image text to text
  • transformers

Tags

  • transformers
  • internvl_chat
  • multimodal
  • vision-language
  • spatial-reasoning
  • spatiolm

Technical specifications

Parameters2.3B
Librarytransformers
LanguagesNot available
Licenseapache-2.0