hugpadhugpad

allura-forge / l4-scout-linearized-nf4-fixed

Vision32B150Updated May 7, 2025
Hugging Face

Downloads

15

Likes

0

Last updated

May 7, 2025

License

llama4

This is a 4-bit Quantized version of this model with the experts broken up and linearized so they play nicely with PEFT/LoRA. To use this with Axolotl, simply include this in your YAML:

Tasks & library

  • image text to text
  • transformers

Tags

  • transformers
  • llama4
  • image-text-to-text
  • facebook
  • meta
  • pytorch

Technical specifications

Parameters32B
Librarytransformers
Languages12 languages
Licensellama4