hugpadhugpad

IFM / Jais-3-MoVA-36B-A4B-FP8

Text Generation37B1.3K0Updated Oct 27, 2026
Hugging Face

Downloads

1.3K

Likes

0

Last updated

Oct 27, 2026

License

apache-2.0

[!NOTE] This repository contains an FP8-quantized version of inception42/Jais-3-MoVA-36B-A4B. Only the routed-expert linear layers in transformer layers 3–47 are quantized to FP8: - Weights: static FP8 E4M3FN, one FP32 scale per 128128 block. - Activations: dynamic FP8, one s…

Tasks & library

  • text generation
  • transformers

Tags

  • transformers
  • k2_horizon
  • text-generation
  • jais-3
  • 36b
  • mova

Technical specifications

Parameters37B
Librarytransformers
Languagesar, en
Base modelinception42/Jais-3-MoVA-36B-A4B
Licenseapache-2.0