Cortex
cortexformac.com- Category
- Productivity
- Rank
- No. 4218Tools index
Previous survey · No. 4016 ·
- Pricing
- Paid
- Type
- APP
- Use case
- Deployment & Operations
- Date
About
Cortex is a macOS system monitor that groups about a thousand processes into the roughly sixty apps you actually run and keeps 30 days of history. It shows Neural Engine power, detects local AI workloads such as Ollama and LM Studio, and includes a model fit advisor comparing a model's memory needs with free memory. Data stays on the Mac. It requires macOS 15 or later on Apple Silicon.
At a glance
Cortex groups Mac processes by app and shows resource use alongside estimates for local AI models. It is worth a look if you regularly troubleshoot memory pressure or want to estimate whether a model will fit alongside the apps you already have open.
Good fit
Apple Silicon users running local models or diagnosing which app is behind a resource spike.
Consider something else when
If you only need to identify a process using lots of CPU occasionally, start with the Activity Monitor already on your Mac.
Start here
Use the website’s sample-data preview to see how Cortex groups Chrome processes and presents model memory estimates. Treat that preview as a tour. On your own Mac, compare the estimate for a model with memory pressure while it loads.
Cost and requirements
Requires macOS 15 or later. Intel Macs have fewer metrics. The vendor lists a one-time per-Mac license and no trial, with a limited launch price.
Limits to know
Model fit is an estimate that depends on the chosen quantization and context length. The vendor says metrics and history stay on the Mac and describes license and update checks plus an optional Anthropic usage check.
Sources checked 2026-11-05. Source-based assessment.
Why it made the leaderboard
It gives local-AI developers visibility Activity Monitor lacks: Neural Engine power draw, which apps hold model memory, and whether a given quantized model will actually fit before you try loading it.
Intel on Cortex
Tags
Media
Comments (0)
No comments yet
Editorially curated, with community endorsements as a secondary signal. Corrections welcome.