Model Overview

Description

This is a copy of NVIDIA Gemma 4 26B A4B NVFP4, a quantized variant of Google Gemma 4 26B A4B IT.

The intent is to provide NVIDIA's weights with the latest configuration updates from Google applied on top.

Refer to the original model cards linked above for model descriptions, capabilities, benchmarks, usage and limitations.

Downloads last month
210
Safetensors
Model size
14B params
Tensor type
BF16
·
U8
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for contyk/Gemma-4-26B-A4B-NVFP4

Quantized
(336)
this model