GPU comparison
MacBook Pro / Mac mini M4 Pro (24GB) vs Mac mini M6 (16GB) for local LLMs
How the two stack up for running open-source LLMs locally — memory, bandwidth, price, and how many of the 64 tracked models each one can run.
| MacBook Pro / Mac mini M4 Pro (24GB) | Mac mini M6 (16GB) | |
|---|---|---|
| Memory | 24.0 GB | 16.0 GB |
| Bandwidth | 273 GB/s | 153 GB/s |
| Price (approx) | $1,999 | $899 |
| LLMs it runs | 45 of 64 | 27 of 64 |
| Best model it runs | Qwen3.8 27B · 12–19 tok/s | Devstral Small 2 24B · 10–15 tok/s |
The MacBook Pro / Mac mini M4 Pro (24GB) runs 18 more of the tracked models (45 vs 27), thanks to its 24.0 GB of memory. The MacBook Pro / Mac mini M4 Pro (24GB) has more memory bandwidth (273 GB/s), so it generates tokens faster at the same model and quant.