Can I Run Mistral Small 3.2 on 32 GB system RAM?

Written by Jakub Rusinowski · Last updated September 6, 2026

Yes — comfortably

Yes, comfortably — Mistral Small 3.2 24B at Q3_K_M needs about 12.5 GB of the 25.6 GB usable on 32 GB system RAM, leaving ~13.1 GB spare and running at ~6.1 tok/s (estimated), with room for about 65,536 tokens of context.

Confidence: medium · Recommended quantization: Q3_K_M · Estimated speed: ~6.1 tok/s

Buy This HardwareAMD Radeon RX 9060 XT 16GB — 16 GB VRAM · 160 W board power Deploy in the Cloud NowRTX 4090 on RunPod — from $0.34/hr · rate checked 2026-07

or compare on Vast.ai from $0.35/hr (typical low · varies)

As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.

32 GB system RAM — what it gives a model

Usable memory for models	25.6 GB
Memory bandwidth	90 GB/s

Mistral Small 3.2 on 32 GB system RAM: memory by quantization

Quant	Memory needed	Fits 25.6 GB?	Max context	Est. speed	Download
F16	49.6 GB	✗ No	—	—	47.2 GB
Q8_0	27.5 GB	✗ No	—	—	25.1 GB
Q6_K	21.8 GB	✓ Yes	16K	~3.3 tok/s	19.4 GB
Q5_K_M	19.2 GB	✓ Yes	32K	~3.8 tok/s	16.7 GB
Q4_K_M	16.7 GB	✓ Yes	32K	~4.4 tok/s	14.2 GB
Q3_K_M	12.5 GB	✓ Yes	64K	~6.1 tok/s	10.1 GB
Q2_K	10.2 GB	✓ Yes	64K	~7.7 tok/s	7.8 GB

What to watch out for

Q3_K_M is the only quantization that fits, and it is a heavily degraded one — expect noticeably worse output than the same model at Q4_K_M. A smaller model at Q4 is usually the better trade.
This model does not publish its full attention configuration, so the KV-cache share of these figures is inferred from its parameter count rather than computed exactly.
These figures assume CPU-only inference. Any discrete GPU, even an 8 GB one, will be several times faster for models that fit in its VRAM.

Recommended setup

llama.cpp (CPU build) or Ollama — both run without a GPU

How these numbers are calculated

Memory need = quantized weights + KV cache at 8,192 tokens (f16) + 0.8 GB runtime overhead.
25.6 GB of the 32 GB is treated as usable for model weights (80% — the rest is the OS and running applications).
DDR5-5600 dual channel at 89.6 GB/s peak. CPU decode is assumed to sustain 35% of that peak, because CPU inference is not purely bandwidth-bound — it also spends real time in compute and thread synchronisation. This figure is an assumption, not a fitted constant: no CPU measurement is in the calibration set.
CPU-only inference: no GPU is assumed. A GPU of any size will beat these figures substantially.
Throughput is a memory-bandwidth roofline estimate, not a measurement. It is labelled "estimated" everywhere it appears.
KV cache is inferred from the parameter count because this model does not publish its full attention configuration.

FAQ