Can I Run Mistral Family on Mac mini (M4 Pro, 64 GB)?

Superseded model. Mistral Family has been superseded by Mistral Small 4. This page is kept for reference; the newer family is a better starting point. View Mistral Small 4 →

Written by Jakub Rusinowski · Last updated January 28, 2025

Yes — comfortably

Yes, comfortably — Mistral Small 3 (24B) at Q8_0 needs about 28.2 GB of the 48 GB usable on Mac mini (M4 Pro, 64 GB), leaving ~19.8 GB spare and running at ~5.6 tok/s (estimated), with room for about 32,768 tokens of context.

Confidence: high · Recommended quantization: Q8_0 · Estimated speed: ~5.6 tok/s

Mac mini (M4 Pro, 64 GB) — what it gives a model

Usable memory for models	48 GB
Memory bandwidth	273 GB/s
Form factor	Mini PC
Operating system	macOS
Memory upgradeable	No — soldered

Mistral Family on Mac mini (M4 Pro, 64 GB): memory by quantization

Quant	Memory needed	Fits 48 GB?	Max context	Est. speed	Download
F16	50.7 GB	✗ No	—	—	48 GB
Q8_0	28.2 GB	✓ Yes	32K	~5.6 tok/s	25.5 GB
Q6_K	22.4 GB	✓ Yes	32K	~7.1 tok/s	19.7 GB
Q5_K_M	19.7 GB	✓ Yes	32K	~8.1 tok/s	17 GB
Q4_K_M	17.2 GB	✓ Yes	32K	~9.4 tok/s	14.5 GB
Q3_K_M	12.9 GB	✓ Yes	32K	~12.9 tok/s	10.2 GB
Q2_K	10.6 GB	✓ Yes	32K	~16.1 tok/s	7.9 GB

Which Mistral Family sizes fit

Variant	Needs at Q4_K_M	Fits?	Est. speed
Mistral Small 3 (24B)	17.2 GB	✓ Fits	~9.4 tok/s
Mistral NeMo 12B	9.4 GB	✓ Fits	~17.9 tok/s

What to watch out for

Memory on this machine is not upgradeable, so the configuration you buy is the ceiling for every model you will ever run on it.

Mac mini M4 Pro 64 GB limitations

M4 Pro memory bandwidth is half the M4 Max's, so large models that fit will still generate roughly half as fast.

Recommended setup

Ollama or LM Studio (Metal) — MLX for the fastest Apple-native throughput

How these numbers are calculated

Memory need = quantized weights + KV cache at 8,192 tokens (f16) + 0.8 GB runtime overhead.
64 GB unified memory at 273 GB/s, shared between CPU and GPU.
macOS reserves a share of unified memory for the system, so not all of it is available to a model.
Throughput is a memory-bandwidth roofline estimate, not a measurement. It is labelled "estimated" everywhere it appears.
KV cache is computed from this model's published attention configuration.

FAQ