News:

Willkommen im Notebookcheck.com Forum! Hier können Sie über alle unsere Artikel und allgemein über notebookrelevante Dinge diskutieren. Viel Spass!

Main Menu

Compact powerhouse with the 2 nm M6 SoC - Apple Mac mini 2026 Review

Started by Redaktion, September 21, 2026, 15:01:31

Previous topic - Next topic

Redaktion

Apple now offers the Mac mini with the new M6 SoC, their first 2 nm chip. With significant CPU and GPU performance improvements, the small desktop system packs quite a punch and easily beats the rivals from AMD and Intel.

https://www.notebookcheck.net/Compact-powerhouse-with-the-2-nm-M6-SoC-Apple-Mac-mini-2026-Review.1401148.0.html


Lawrence

There is no microphone and asking SIRI AI questions is cumbersome. USB C to A adaptors often report disconnects.

hopefully M7 with ~40 GB

QuoteAnd if the M6 is not powerful enough or you need more than 32 GB RAM, Apple also offers the Mac mini with the M5 Pro SoC.
This is good I guess, but the not-improvement from previous M5 to more than 32 GB RAM/unified memory is a disappointment, because if one wants to run SOTA model huggingface.co/unsloth/Qwen3.6-35B-A3B-MTP-GGUF/blob/main/Qwen3.6-35B-A3B-UD-Q4_K_XL.gguf[1] -- and this model would run fast enough on this mini-PC, if it only had 8 GB more RAM/unified memory --, this is how many context tokens one can fit (for this MTP variant, context will be less, because the MTP head also consumes memory):
Quote from: reddit.com/r/LocalLLaMA/comments/1sq94qx/is_anyone_getting_real_coding_work_done_with.. I've come to the conclusion that (1) 32768 is the biggest context I can get away with in an adequately smart model, and (2) it just ain't enough.

Calculating available context: 16,384 [KV cache context tokens per GB]*(32 [GB total memory] - 6 to 8 [GB for the OS] - [quant filesize])[2]:
For huggingface.co/unsloth/Qwen3.6-35B-A3B-MTP-GGUF -> UD-Q4_K_XL.gguf:
~34,400 context tokens = 16,384*(32-7-22.9) (confirms the quote)

Hopefully the M7 will come with more than up to 32 GB RAM, 40 GB would be a good start, if APPLE considers 48 GB to eat too much into M Pro.

[1] artificialanalysis.ai/?models=qwen3-8-27b%2Cqwen3-8-27b-medium%2Cqwen3-8-27b-low%2Cqwen3-8-27b-non-reasoning%2Cqwen3-6-35b-a3b%2Cqwen3-6-35b-a3b-non-reasoning&intelligence=artificial-analysis-intelligence-index
[2] reddit.com/r/Qwen_AI/comments/1vo8pjz/qwen3827b_kv_cache_works_out_to_64_kibtoken_so/

Rolando


Quick Reply

Name:
Email:
Verification:
Please leave this box empty:
Shortcuts: ALT+S post or ALT+P preview