Model•active

Gemma 4 26B A4B

Gemma 4 26B A4B is an open-weight mixture-of-experts model from Google DeepMind with 25.2B total parameters and 3.8B active parameters.

Official siteVerified Oct 4, 2026

Overview

The model supports text and image input, a 256K context window, reasoning, coding and function calling while activating only a fraction of its total parameters during inference.

Model specifications

Family: Gemma 4
Context: 256,000 tokens
Input: Text, Image
Output: Text
Open weights: Yes
Capabilities: Reasoning, Vision, Tool Use, Function Calling

Related entities

Belongs To

Model Familyactive

Gemma 4

Gemma 4 is Google DeepMind's current open-weight multimodal model family, spanning edge-focused models, dense local models and a mixture-of-experts variant.

Related by Develops

Companyactive

Google

Google develops internet, cloud, developer, and AI products, including the Gemini model and developer-tool product family.