No — not on this device
Devstral Small 2 24B at Q4_K_M needs 15.3 GB against 10.7 GB usable, and the shortfall of 4.6 GB is more than 32 GB of system RAM can cover at a tolerable speed. A smaller sibling or a lower quantisation is the honest answer here.
Does not fit
8K context
Q4_K_M · 13.5 GB
Apache 2.0
Released Dec 2025
Vision
Not in the Ollama library
Built for software-engineering agents (OpenHands, Cline). Dense 24B, 384K window. Not in the Ollama library.
The VRAM budget
weights 13.5 GB
Weights 13.5 GB
KV cache @ 8K 1.25 GB
Runtime overhead 0.6 GB
Over budget 4.6 GB past 10.7 GB
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| Q8_0 | 23.7 GB | 25.6 GB | — | ~1.6 | −0.1% ppl | 14.9 GB over |
| Q6_K | 18.3 GB | 20.2 GB | — | ~2.1 | −0.4% ppl | 9.5 GB over |
| Q5_K_M | 15.8 GB | 17.7 GB | — | ~2.4 | −0.8% ppl | 7.0 GB over |
| Q4_K_M | 13.5 GB | 15.3 GB | — | ~2.8 | −1.9% ppl | 4.6 GB over |
| Q3_K_M | 10.9 GB | 12.8 GB | — | ~3.5 | −5.4% ppl | 2.1 GB over |
| Q2_K | 9.4 GB | 11.2 GB | 4K | ~4.1 | −15% ppl | 0.5 GB over |
Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it.
How to run it
$ pip install mlx-lm $ mlx_lm.generate --model mlx-community/Devstral-Small-2-24B-Instruct-2512-4bit \ --max-tokens 512 --prompt "Hello"
Apple's own array framework. The fastest path on Apple Silicon. More on MLX.
01Download is 13.5 GB. Keep it on an SSD — a first load off a spinning disk takes minutes.
02macOS caps what the GPU may wire down at about 10.7 GB of 16 GB. Raising it with iogpu.wired_limit_mb is possible, and risky.
03If it falls back to CPU silently, drop the context first, then step down a quantisation.