Gemma 4

Google

Gemma 4 26B A4B (Non-reasoning) is a 26-billion parameter open model from Google, optimized for general-purpose tasks without specialized reasoning enhancements. It is part of the Gemma family, built on Gemini technology, and designed for efficient performance across a wide range of applications.

Google

Gemma 4 26B A4B is a 26-billion parameter reasoning model from Google's Gemma series. It is designed for complex reasoning tasks and supports multimodal inputs, making it suitable for advanced analysis and problem-solving.

Google

Gemma 4 31B (Non-reasoning) is a 31-billion parameter open model from Google, optimized for general-purpose tasks and fast inference. It is part of the Gemma family, derived from Gemini technology, and is designed for efficient performance in coding, instruction following, and multimodal understanding without the overhead of extended reasoning chains.

Google

Gemma 4 31B (Reasoning) is a reasoning-optimized variant within Google's Gemma model family. It is designed to excel at tasks requiring deep, step-by-step thinking and logical deduction, such as complex problem-solving and analysis.

Google

Gemma 4 E2B is a lightweight, efficient model from Google's Gemma family, optimized for fast inference on edge devices or in-browser environments. As a non-reasoning variant, it prioritizes speed and low resource consumption for straightforward tasks.

Google

Gemma 4 E2B (Reasoning) is a lightweight, open-source model from Google optimized for efficient reasoning tasks. It is designed to deliver strong performance on logic, math, and code-related problems while maintaining a small footprint suitable for edge or resource-constrained deployments.

Google

Gemma 4 E4B (Non-reasoning) is a lightweight, efficient 4B parameter model from Google's Gemma series, optimized for fast inference and low-cost deployment. It is designed for general-purpose tasks where speed and efficiency are prioritized over complex reasoning chains.

Google

Gemma 4 E4B (Reasoning) is a lightweight, 4-billion parameter model from Google's Gemma series, optimized for strong reasoning and instruction-following capabilities. It is designed for efficient deployment while maintaining performance on complex tasks, and supports multimodal inputs.