HARDWARE / UNIFIED MEMORY

Mac mini M6 · 24 GB

A compact current-generation Mac with 24 GB unified memory. Keep macOS, runtime, and KV-cache headroom separate from model weights.

Primärquelle

Current compact Apple desktop24 GB

01 / PRACTICAL CEILING

Qwen3.6 27B

The highest listed result may be a tight or offloaded fit. For daily use, prefer the first result marked Comfortable and keep context modest.

02 / DEFAULT RUNTIME

MLX LM

Native Apple silicon inference, experimentation and fine-tuning. Only for Apple silicon. Use an MLX-converted model and tune KV cache size for long contexts.

Leitfaden öffnen

03 / MODEL ENVELOPE

Repräsentative Open-Weight-Modelle

Alibaba Qwen

Qwen3.6 27B

27BQ4 / NVFP4 estimate

Geschätzter Speicher: ~18.5 GB
Kontext: 262K native

Mistral AI

Devstral Small 2 24B

24BGGUF Q4 / BF16

Geschätzter Speicher: ~18.5 GB
Kontext: 256K

OpenAI

gpt-oss-20b

21B3.6B active · MXFP4

Geschätzter Speicher: ~16 GB
Kontext: 128K

Alibaba Qwen

Qwen3 14B

14BGGUF Q4

Geschätzter Speicher: ~11.2 GB
Kontext: 32K+

Google

Gemma 3 12B

12BINT4

Geschätzter Speicher: ~9.4 GB
Kontext: 128K

Alibaba Qwen

Qwen3 8B

8BGGUF Q4

Geschätzter Speicher: ~6.8 GB
Kontext: 32K+

Achtung

Fit is based on estimated total model memory. Driver support, KV cache, multimodal projectors, concurrency, and desktop applications can all reduce available headroom. Current fit: Qwen3.6 27B (Tight), Devstral Small 2 24B (Tight), gpt-oss-20b (Tight), Qwen3 14B (Comfortable), Gemma 3 12B (Comfortable), Qwen3 8B (Comfortable).