I tried converting it as a test, and surprisingly, it works. Maybe because of a bug, the --jinja parameter doesn't work.

GGUF

Model size

27B params

Architecture

gemma3

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for yeoj34760/translategemma-27b-it-MXFP4_MOE-GGUF

Base model

Quantized

(15)

this model