Apple · Apple silicon · 2026
128 GB at 614 GB/s: gpt-oss-120b and 122B MoE models at a good pace, almost silently.
from $2,499 (launch price)
Researched from manufacturer data and public sources; not tested by us.· Last reviewed Oct 11, 2026
Popular models at Q4_K_M with a 32K-token context. Memory needed from real model files; speed is the theoretical ceiling.
Model weightsContext cache (KV)RuntimeUsable limit (colour = verdict)
| Model | Verdict | Memory | Speed ceiling |
|---|---|---|---|
| Qwen3 1.7B | Runs | ≤ 414 tok/s | |
| Qwen3.5 9B | Runs | ≤ 81 tok/s | |
| Gemma 4 12B | Runs | ≤ 65 tok/s | |
| gpt-oss-20b | Runs | ≤ 230 tok/s | |
| Qwen3.6 27B | Runs | ≤ 27 tok/s | |
| Qwen3.6 35B A3B | Runs | ≤ 333 tok/s | |
| Llama 3.3 70B | Runs | ≤ 14 tok/s | |
| gpt-oss-120b | Runs | ≤ 224 tok/s | |
| Qwen3.5 122B A10B | Runs | ≤ 100 tok/s | |
| Qwen3.5 397B A17B | Not enough memory | — |
Theoretical maximum for generating text: memory bandwidth divided by the bytes read per token. Real runs usually reach about half to three quarters of it.
Prices are editorial references (launch price and a typical street price on the date shown), not live prices. In 2026 the memory shortage has pushed many prices well above launch.
Prices are shown only if updated in the last 24 hours and may change; the retailer’s price at checkout applies. We may earn a commission from these links at no cost to you. Affiliate disclosure
Best for: The best all-round local AI machine for most enthusiasts with a serious budget.
Yes with 128 GB: our top pick for 100B-class MoE models in one quiet box.
macOS lets the GPU use about two thirds of memory on machines with up to 36 GB, and about three quarters above that. The limit can be raised (sudo sysctl iogpu.wired_limit_mb=…), leaving less for macOS and other apps; our figures use the default. The base model (32-core GPU) has 460 GB/s; the 40-core GPU version 614 GB/s.
Source: Manufacturer specifications
How well the popular local-AI tools run on it today (editorial assessment).
Measured results with their source. We never invent speeds.
No measured results yet. The table above shows the theoretical ceiling instead.
Ryzen AI
Largest popular model at 4-bit: Qwen3.5 122B A10B
128 GB of unified memory in a mini PC for less than a Mac Studio: the homelab favourite for MoE models.
Same 128 GB for less, on Linux