NVIDIA RTX and Apple Silicon macOS Powerful Pairing for AI
eGPU connections were a definite shortcoming for MacBook users, especially now with the new M-series chips. That's why TinyCorp has released a proprietary driver, which does the work of pinning NVIDIA GeForce RTX boards to any new mac.
The Big Reservation For AI, not for Gaming
At this stage, however, it is essential to be explicit about the limitations of this workaround. The connected NVIDIA GPU is a pure AI computational accelerator. It means it does not output video, nor can it run games, render graphics, or even accelerate the macOS interface. The only function assigned is to speed up operations on large language models (LLMs) and neural networks in major proportions.
How It Works and Who It Is Intended For
This is a breakthrough for AI developers and tech-savvy enthusiasts alike. Mostly not plug-and-play, this will require some technical know-how as the user will have to install the drivers manually and turn off SIP (System Integrity Protection) core security features of macOS.
This new driver follows TrueCorp's previous success in allowing external AMD GPU support for Apple Silicon Macs. The tests have been successfully carried out on a MacBook Pro with an M3 Max chip by connecting the external RTX card, as the ADT-UT3G dock via USB 4 or Thunderbolt 4.
NVIDIA over USB4 on MacBook is ready to try!
— the tiny corp (@__tinygrad__) October 20, 2025
* ADT-UT3G dock + any 30/40/50 series GPU
* Disable SIP
* Install driver `extra/usbgpu/tbgpu`
* Install NVK compiler `brew install tinymesa`
* Test with:
`DEBUG=2 NV_NAK=1 NV=1 python3 test/test_tiny.py TestTiny.test_plus` pic.twitter.com/bWVVmC4x8E
Supported Graphics Cards
It comes under the very selective eligibility for this solution that has been set for certain generations of NVIDIA cards:
- Full Support: NVIDIA GeForce RTX 30, 40, and 50 series.
- Partial Support: Some NVIDIA GeForce RTX 20 series cards may have limited compatibility.
- Not Supported: NVIDIA GTX series graphics cards.
The Primary Advantage
The key advantage of this development is that it allows AI experts to use the incredible computing capacities of NVIDIA RTX without abandoning macOS. No requirement for switching to a Windows or Linux machine for training and running AI models, making it easier for developers who prefer to continue their jobs on Mac computers.
