In addition to Google AI Edge Gallery, which allows users to run Gemma models locally on their Macs, the company also launched the Gemma 4 12B model and the Google AI Edge Eloquent dictation app for Mac. Here are the details.
A little background
Most users who rely on LLMs for everyday tasks tend to use ChatGPT, Claude, or Gemini, which are cloud-based models running on OpenAI, Anthropic, and Google servers.
Another way to interact with LLMs is through local models. They tend to be much smaller and less capable than the trillion-parameter models that run in the cloud, but they also have several advantages.
On the one hand, being less capable than cloud-based models does not mean they are bad. Additionally, they do not require an active Internet connection, as they run on the computer’s own processing power. Additionally, the better the computer, the faster the responses and the larger the models it can handle. And finally, because everything runs locally, these models are also more private, since the conversation data doesn’t need to leave the device.
There are a few ways to install local models on a Mac and we cover this herewhen OpenAI launched its own open models. But in a nutshell, you need to install platforms like Be and LM Studioand then install a model that can run smoothly on your Mac hardware.
hugging face hosts thousands of open models to choose from, including those from frontier laboratories. However, platforms like Ollama and LM Studio also offer ways to install these models directly from them.
Which brings us to Google AI Edge GalleryGoogle’s platform for running AI models locally. Google already offered a Google AI Edge Gallery app for Android and iOS, but today the company also launched it for macOS.
Google AI Edge and Gemma 4 12B Gallery
One thing to note from the start is that unlike Ollama and LM Studio, which allow users to install any AI model supported by their hardware, Google AI Edge Gallery for Mac currently only offers access to 5 of Google’s own models, where ‘it’ stands for instruction, meaning they can be adjusted to follow the user’s instructions rather than simply filling in the text:
- Gemma-4-12B-it
- Gemma-4-E2B-it
- Gemma-4-E4B-it
- Gemma-3n-E2B-it
- Gemma-3n-E4B-it
The item at the top of the list is particularly notable. Gem 4 12B “launched today and was designed to bring multimodal and agent intelligence right to your laptop,” according to Google.
While most consumer-oriented on-premise models from cutting-edge AI labs tend to stay between 2 billion and 9 billion parameters, Google says Gemma 4’s 12 billion-parameter design offers performance comparable to its 26 billion-parameter expert mixing model, while still being “small enough to run locally on consumer laptops with 16GB of RAM.”
Gemma 4 12B is also multi-modal, meaning it can handle text, vision and audio. Google says the model also includes good encryption capabilities, “allowing you to extract meaningful insights from your data right on your device.”
You can learn more about Google AI Edge Gallery hereand you can learn more about Gemma 4 12B here.
Google AI Edge Eloquent
In addition to Gemma 12B and the launch of Google AI Edge Gallery for macOS, Google also launched Google AI Edge Eloquent app for Mac today, after bringing the app to iOS a few months ago.
Google AI Edge Eloquent is a free dictation app that captures what users say and transcribes it while polishing text, removing disfluencies, and making light edits for clarity and fluency. Processing is done on the device, rather than in the cloud.
The app also allows users to choose from different writing styles and add custom words, such as names, slang, and other terms they use frequently. That helps avoid the kind of frequent miscorrections that dictation apps can make with specific words and phrases.
You can learn more about Google AI Edge Eloquent here.
Worth checking out on Amazon
FTC: We use automatic affiliate links that generate income. Further.
