Hugging Face | GitHub | Launch Blog | Documentation
License: Apache 2.0 | Authors: Google DeepMind

This repository contains a FP8 dynamic version of gemma-4-12B-it, quantized to run on <=16GB VRAM hardware. Given the quantization method, it requires Ada-based or later architectures.

Downloads last month: 16

Safetensors

Model size

12B params

Tensor type

BF16

F8_E4M3

Inference Providers NEW

Any-to-Any

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

Model tree for edornd/gemma-4-12B-it-FP8D

Base model

google/gemma-4-12B

Finetuned

google/gemma-4-12B-it

Quantized

(404)

this model