hugpadhugpad

ibm-granite / granite-docling-258M

Vision258M125.1K1.3KUpdated Sep 23, 2025
Hugging Face

Downloads

125.1K

Likes

1.3K

Last updated

Sep 23, 2025

License

apache-2.0

Granite Docling 258M builds upon the Idefics3 architecture, but introduces two key modifications: it replaces the vision encoder with siglip2-base-patch16-512 and substitutes the language model with a Granite 165M LLM. Try out our Granite-Docling-258 demo today.

Tasks & library

  • image text to text
  • transformers

Tags

  • transformers
  • idefics3
  • image-text-to-text
  • text-generation
  • documents
  • code

Technical specifications

Parameters258M
Librarytransformers
Languagesen
Licenseapache-2.0