Can I Run Devstral Small 2 24B Instruct 2512 on a Apple M3 Pro (36GB)?
Runs at Q8_0 — near-lossless quality.
12 quantizations fit your 36.0GB
| Quant | Min VRAM | Recommended | File size | Headroom |
|---|---|---|---|---|
| Q8_0BEST | 26.5 GB | 28.0 GB | 25.1 GB | +9.5 GB |
| Q6_K | 20.8 GB | 22.3 GB | 19.4 GB | +15.2 GB |
| Q5_K_M | 18.0 GB | 19.5 GB | 16.8 GB | +18.0 GB |
| Q5_K_S | 17.6 GB | 19.1 GB | 16.3 GB | +18.4 GB |
| Q4_1 | 16.0 GB | 17.5 GB | 14.9 GB | +20.0 GB |
| Q4_K_M | 15.6 GB | 17.1 GB | 14.3 GB | +20.4 GB |
| Q4_K_S | 14.7 GB | 16.2 GB | 13.6 GB | +21.3 GB |
| Q4_0 | 14.5 GB | 16.0 GB | 13.5 GB | +21.5 GB |
| Q3_K_L | 11.7 GB | 13.2 GB | 12.4 GB | +24.3 GB |
| Q3_K_M | 11.1 GB | 12.6 GB | 11.5 GB | +24.9 GB |
| Q3_K_S | 10.2 GB | 11.7 GB | 10.4 GB | +25.8 GB |
| Q2_K | 8.9 GB | 10.4 GB | 8.9 GB | +27.1 GB |
Try it in the cloud first
Don't want to download Devstral Small 2 24B Instruct 2512 just to try it? Use a hosted API or rent a GPU by the second.
Affiliate links — we earn a commission at no cost to you.
All quant variants, benchmark scores, and use-case tags.
Top-ranked open-source models that fit in 36.0GB.
FAQ
Can the Apple M3 Pro (36GB) run Devstral Small 2 24B Instruct 2512?
Yes. The Apple M3 Pro (36GB)'s 36.0GB of unified memory is enough to run Devstral Small 2 24B Instruct 2512 at Q8_0 quantization (26.5GB required).
What's the best quantization to use?
Q8_0 is the highest-precision quantization that fits in your 36.0GB. It uses about 26.5GB of memory and 28.0GB recommended for comfortable inference.
What if I need more headroom for context length?
KV cache memory grows with context length. The numbers above assume a baseline 2K-4K context. For long-context use (32K+), add another 2-6GB depending on the model architecture.