# Google Releases Gemma 4, a 12B 'Any-to-Any' Model

> The new 12-billion-parameter model from Google DeepMind is designed to handle a flexible mix of data types, moving beyond traditional text and image inputs.

Published by The Open Weights on May 23, 2026. Canonical: https://theopenweights.com/news/gemma-4-12b-lou5

## Key facts

- Company: Google DeepMind
- Model: Gemma 4
- Version: 4-12B-it
- Category: Any-to-Any
- Modalities: Any-to-Any, Vision-Language, Text / LLM
- License: Gemma (Open weights, commercial use allowed)
- Parameters: 12B
- Context window: 262K tokens
- Architecture: Gemma4UnifiedForConditionalGeneration
- Precision: BF16
- Weights size: 23.9 GB
- Significance: major
- Published: 2026-05-23
- Last verified: 2026-11-01
- Hugging Face: https://huggingface.co/google/gemma-4-12B-it
- Canonical URL: https://theopenweights.com/news/gemma-4-12b-lou5

Google DeepMind has expanded its open-weights portfolio with the release of Gemma 4 12B Instruct, a new 12-billion-parameter model. The model's key innovation is its unified 'any-to-any' multimodal architecture, designed to handle diverse data inputs and outputs seamlessly.

Unlike traditional models that are often limited to specific input-output pairs like text-to-image, Gemma 4 is built for more flexible, generalized reasoning. According to the [release details on Hugging Face](https://huggingface.co/google/gemma-4-12B-it), its 'any-to-any' design allows it to process combinations of modalities simultaneously, a significant step toward more capable AI systems.

## Why It Matters

The arrival of Gemma 4 democratizes a sophisticated architecture previously seen in much larger, closed models. By packaging these capabilities into a relatively efficient 12B parameter model, Google enables a wider range of researchers and developers to experiment with advanced multimodal applications that require less computational overhead.

The model is available under the custom Gemma license, which includes specific terms for usage and distribution. As an instruction-tuned variant, Gemma 4 12B is optimized for direct use in conversational and task-oriented applications.

## Get the model

- [Hugging Face](https://huggingface.co/google/gemma-4-12B-it)
- [GitHub](https://github.com/google-deepmind/gemma/releases/tag/v4.0.0)

## Sources

- [google/gemma-4-12B-it](https://huggingface.co/google/gemma-4-12B-it) — Hugging Face, May 23, 2026
- [google-deepmind/gemma v4.0.0](https://github.com/google-deepmind/gemma/releases/tag/v4.0.0) — GitHub, May 13, 2026

---
Source: The Open Weights (https://theopenweights.com/). Aggregated and written by Claude, curated by humans. Cite as: "Google Releases Gemma 4, a 12B 'Any-to-Any' Model", The Open Weights, May 23, 2026, https://theopenweights.com/news/gemma-4-12b-lou5