The GPU tab
The GPU tab shows how busy the GPU is, its memory, the average and the peak, and the apps using it with the heaviest first. A render, a game or a browser tab shows up under its own name. The chart covers the last 5, 15 or 30 minutes.
A card names the top app, and the list under it gives each app’s share of the GPU with its process count.
The GPU’s temperature is with the other sensors. The GPU is also in the history, so a render from an hour ago is still there to see. Activity Monitor has a GPU column and a GPU History window.
A number for the Neural Engine
On Apple silicon the Neural Engine does real work, for Photos, Spotlight, dictation and on-device models, and Activity Monitor shows nothing for it. Vitals works out how hard it is working from the memory it moves, against the most this Mac has ever moved.
macOS does not say which app is working the Neural Engine, only which apps hold it open. Those are the apps Vitals names. The tab exists on Macs with Apple silicon. An Intel Mac has no Neural Engine and no tab.
The tab shows the percentage with its average and peak, the memory traffic behind the figure, and each app with the number of connections it holds. Work that used to be invisible, such as Photos indexing a library overnight, now has a name and a chart.
Local models run on the GPU
Models run with PyTorch or MLX use the GPU, not the Neural Engine. An Ollama session therefore shows as a busy GPU and a quiet Neural Engine. That is correct, and surprising the first time. Ollama’s memory is listed under Ollama, where you would look for it.
So if you are checking whether a local model is using the chip, look at the GPU tab. A quiet Neural Engine during a model run does not mean something is wrong.
Menu bar, Overview and notices
Both can be menu bar items, as a number, a graph or both, in whatever order you like, and each has a card on the Overview. That is enough to keep an eye on a long render without opening anything.
The GPU reading turns red at 90% in the menu bar. Vitals also sends a notice when an app has kept the Neural Engine busy for twenty minutes.
How to use it
- 1
Open the GPU tab
See how busy the GPU is, its memory, the average and peak, and the apps using it.
- 2
Open the Neural Engine tab
On a Mac with Apple silicon, see how hard it is working and which apps hold it open.
- 3
Add them to the menu bar
In Settings → Layout → Menu Bar, add a GPU item and a Neural Engine item as a number or a graph.
The details
- GPU tab
- Use, memory, average and peak, with the apps using it
- Neural Engine tab
- How hard it is working and the apps that hold it open, on Apple silicon
- Charts
- The last 5, 15 or 30 minutes
- How the Neural Engine is measured
- The memory it moves, against the most this Mac has moved
- Menu bar
- An item for each; the GPU turns red at 90%
- Notice
- When an app keeps the Neural Engine busy for twenty minutes
Questions
Does Vitals show Neural Engine usage?
Yes, on Macs with Apple silicon: how hard it is working, which apps are using it, and a chart of the last minutes.
Why does my local model not show on the Neural Engine?
Models run with PyTorch or MLX use the GPU, not the Neural Engine. They show on the GPU tab.
How does Vitals measure the Neural Engine?
From the memory it moves, compared with the most this Mac has ever moved. macOS does not say which app is working it, only which apps hold it open, and those are the ones Vitals names.
Does it show GPU memory?
Yes, with the GPU’s use, average and peak, and the apps using it.
Does Activity Monitor show this?
It has a GPU column and a GPU History window, and nothing for the Neural Engine.
The same feature on the app's own site: vitalsmac.com/features/gpu-and-neural-engine