Buying a MacBook Air M5 with 24GB of RAM is recommended for users studying AI or running local Large Language Models (LLMs), as these tasks require significant memory to load models efficiently. According to Apple specialist Pedro Aznar, director of Applesfera, the M5’s improved per-core AI accelerators make higher RAM capacities a good idea for future-proofing hardware against evolving local AI demands.
Why 24GB of RAM Matters for Local AI on MacBook Air M5
Memory is an important feature when buying a new computer. While 16GB of RAM may not perform poorly, loading LLMs directly onto a machine requires extra memory to avoid system slowdowns.
Pedro Aznar, director of Applesfera, states that users should opt for 24GB of RAM because AI will advance significantly at the local level. He notes that this extra memory is almost certainly required to load LLMs on a Mac, making the upgrade a strategic move for those wanting a “future-proof” device.
Pro Tip: If you are pursuing a degree in AI, prioritize RAM.
Apple Silicon: Unified Memory vs. Traditional RAM
Unlike Windows, which is an operating system that must work on all types of computers from all types of manufacturers, macOS is designed only for those manufactured by Apple. This makes it more optimized, and the RAM and hardware needs in general are different.
This optimization means macOS often requires different RAM than Windows. However, AI workloads change this math. When a local LLM is loaded, it occupies a portion of this memory.
Comparing 16GB vs. 24GB for AI Workloads
| Feature | 16GB Configuration | 24GB Configuration |
|---|---|---|
| General Use | Smooth performance | Higher capacity |
| Local LLMs | Base model | Extra memory for loading LLMs |
| Future-Proofing | Standard | High (AI-ready) |
The Role of M5 AI Accelerators
The hardware shift isn’t just about the amount of memory, but how it’s used. Aznar highlights that the MacBook Air M5 is an excellent choice for AI because its per-core AI accelerators are much better than in the previous generation for those tasks.
When paired with 24GB of RAM, the M5 can process tasks more efficiently.
Did you know? Local AI execution means processing occurs on your device.
Frequently Asked Questions
Is 16GB of RAM enough for a Mac Mini M4?
For standard tasks, yes. One user reports that a Mac Mini M4 with 16GB “va como la seda” (works like a charm), but expresses concern about how it may age with AI.
Can I upgrade the RAM on my MacBook Air later?
The sources do not provide information on whether RAM can be upgraded after purchase.
Why is 24GB specifically recommended for AI students?
AI models (LLMs) are loaded into the RAM. More RAM allows the computer to hold the memory required for these models.
Are you planning your next Mac upgrade? Tell us in the comments if you’re prioritizing RAM or storage for your AI workflow, or subscribe to our newsletter for more hardware analysis.
Related reading
