In addition to the Google AI Edge Gallery, which lets users manage Gemma models natively on their Macs, the company also released the Gemma 4 12B model for Mac and the Google AI Edge Eloquent dictation software. Here are the details.
A little background
Most users who rely on LLMs for day-to-day 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 LLM is through local models. These are typically smaller and less capable than trillion-parameter models running in the cloud, but they also have a number of advantages.
First, just because they are less capable than cloud-based models doesn’t mean they are bad. Also, they do not require an active internet connection because they run on the computer’s own processing power. Also, the better the computer, the faster the answers and the bigger the models it can handle. And finally, because everything works locally, these models are also more private, since chat data does not need to leave the device.
There are several ways to install native models on a Mac and we covered it hereWhen OpenAI released its open models. But in short, what platforms should be installed to be and LM Studioand then install a model that can run smoothly on your Mac’s hardware.
Hugging Face hosts thousands of open models to choose from, including those from frontier labs. However, platforms like Ollama and LM Studio also offer ways to install these models directly from them.
Which brings us Google AI Edge GalleryGoogle’s platform for running AI models natively. Google already offered the Google AI Edge Gallery app for Android and iOS, but today the company released it for macOS as well.
Google AI Edge Gallery and Gemma 4 12B
One thing to note at the outset is that unlike Ollama and LM Studio, which allow users to install any AI model suitable for 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 user instructions instead of completing text:
- Gemma-4-12B-this
- Gemma-4-E2B-it
- Gemma-4-E4B-it
- Gemma-3n-E2B-it
- Gemma-3n-E4B-it
The top item on the list is especially noteworthy. Gemma 4 12B It was released today, and it’s designed to bring agentic, multimodal intelligence right to your laptop,” according to Google.
While most consumer-facing native models from frontier AI labs tend to be between 2 billion and 9 billion parameters, Google says the Gemma 4’s 12 billion parameter design provides performance comparable to a 26 billion parameter expert blend model, while being “small enough to run natively on consumer laptops with 1GB of RAM.”
The Gemma 4 12B is also multi-modal, meaning it can handle text, vision and audio. Google says the model also has good coding 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
Along with the release of Gemma 12B and Google AI Edge Gallery for macOS, Google also 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 software that captures and transcribes what users say while polishing text, removing errors, and making light edits for clarity and fluency. Processing is done on the device, not in the cloud.
The app also allows users to choose between different writing styles and add custom words such as names, slang and other terms they use frequently. This helps avoid common mistakes that spelling programs can make with specific words and phrases.
You can learn more about Google AI Edge eloquent here.
It’s worth checking out on Amazon
FTC: We use automatic affiliate links that generate income. More.
