armand0e

gemma-4-26B-opus-v3

README

License: apache-2.0

Gemma 4 26B A4B Opus v3

This is highly experimental. The model was trained with a custom chat template with preserved reasoning_content across turns. It wasn't stable for inference so the chat template was then reverted back to the official template

Developed by: armand0e
License: apache-2.0
Finetuned from model : armand0e/gemma-4-26B-opus-v3

This gemma4 model was trained 2x faster with Unsloth and Huggingface's TRL library.

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Model Details

Model Provider

armand0e

Model Tree

Base

armand0e/gemma-4-26B-opus-v3

Fine-tuned

this model

Input Modalities

TextImage

Output Modalities

Text

Supported Functionality

Dedicated Endpoints

Explore FriendliAI today

Get started Talk to an engineer