Google has expanded its local AI offerings for Mac users with the launch of Google AI Edge Gallery for macOS, the new Gemma 4 12B model, and the Google AI Edge Eloquent dictation app. The release gives users more ways to run AI directly on their devices without relying on cloud servers.
Local AI models continue to attract attention because they offer greater privacy, offline access, and faster performance on capable hardware. Since processing happens on the device itself, conversations and data do not need to leave the computer. That makes local AI an appealing option for users who want more control over their information.
Google AI Edge Gallery Arrives on Mac
Google AI Edge Gallery is now available for macOS after previously launching on Android and iPhone. Unlike platforms such as Ollama and LM Studio that support a wide range of AI models, Google AI Edge Gallery currently focuses on Google’s own Gemma models.
The app currently supports:
- Gemma-4-12B-it
- Gemma-4-E2B-it
- Gemma-4-E4B-it
- Gemma-3n-E2B-it
- Gemma-3n-E4B-it
These instruction-tuned models are designed to follow user prompts and perform tasks directly on supported Mac hardware.
Gemma 4 12B Targets Powerful Local AI
The highlight of the release is Gemma 4 12B, Google’s newest open model designed for local use. According to Google, the 12-billion-parameter model delivers performance comparable to a much larger 26-billion-parameter mixture-of-experts model while remaining small enough to run on consumer laptops with 16GB of RAM.
Gemma 4 12B also supports multimodal capabilities, allowing it to work with text, images, and audio. Google says the model offers strong coding abilities and can help users analyze data without sending information to external servers.
Alongside AI Edge Gallery, Google has also released Google AI Edge Eloquent for macOS. The free app works as an on-device dictation tool that transcribes speech while improving readability by removing pauses, filler words, and other speech disfluencies.
Users can select different writing styles and add custom words, names, and industry-specific terms to improve accuracy. Because all processing happens locally, users can benefit from faster responses and improved privacy compared to cloud-based transcription services.
With these launches, Google is making local AI more accessible for Mac users who want powerful models and AI-powered productivity tools that work directly on their devices.
