docs: add Apple A-chipsets model size estimation guide

[Dhruvizzle101]

Overview

Adds a tutorial guide for Apple Silicon Mac users on how to estimate
a suitable model size when running LLMs with llama.cpp. The guide covers
unified memory architecture, the model size formula, quantization levels,
a Mac RAM lookup table, practical llama.cpp commands, and recommendations
by memory size. Related discussion: #25065

Additional information

This tutorial addresses the TODO item in issue #13523 — "Apple A-chipsets:
how to estimate a suitable model size?" No existing documentation covered
this topic for Mac users.

Requirements

• I have read and agree with the contributing guidelines
• AI usage disclosure: YES — AI was used in an assistive capacity only to help explain concepts. All tutorial content was written independently by the contributor and verified manually on a MacBook Air with Apple Silicon.

[Dhruvizzle101]

Hi @ggerganov! This is my first contribution to llama.cpp. I've added a tutorial guide addressing the Apple A-chipsets model size estimation TODO from issue #13523. It covers unified memory architecture, the model size formula, quantization levels, a Mac RAM lookup table, and practical llama.cpp commands. Would appreciate a review when you have time!

[ggml-gh-bot]

Hi @Dhruvizzle101, thanks for your contribution!

Per our contribution guidelines, the automated PR checker found the following issue(s) that need your attention:

• AI-generated content: This project does not accept PRs, descriptions or commit messages that are fully or predominantly AI-generated. If you have used AI to assist you in writing code, please make sure to disclose that explicitly.

---

Please note that maintainers reserve the right to make final decisions on PRs. If you believe there is a mistake, please comment below.