No — not on this device
Olmo 3.1 32B Instruct at Q4_K_M needs 20.0 GB against 18.0 GB usable, and the shortfall of 2.0 GB is more than 16 GB of system RAM can cover at a tolerable speed. A smaller sibling or a lower quantisation is the honest answer here.
The largest fully open model you can audit end to end. Q4 fits 24 GB, tightly.
What hardware do I need for Olmo 3.1 32B Instruct? →
Fits instead: Olmo 3 7B Instruct (7.2 GB)
The VRAM budget
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| Q8_0 | 31.9 GB | 33.7 GB | — | ~1.8 | −0.1% ppl | 15.7 GB over |
| Q6_K | 24.6 GB | 26.4 GB | — | ~2.3 | −0.4% ppl | 8.4 GB over |
| Q5_K_M | 21.3 GB | 23.1 GB | — | ~2.6 | −0.8% ppl | 5.1 GB over |
| Q4_K_M | 18.1 GB | 20.0 GB | — | ~3.1 | −1.9% ppl | 2.0 GB over |
| Q3_K_M | 14.7 GB | 16.5 GB | 31K | 3.8 | −5.4% ppl | Long context |
| Q2_K | 12.6 GB | 14.4 GB | 64K | 4.4 | −15% ppl | Long context |
Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it. This model interleaves sliding-window layers (4096 tokens, 16 of 64 layers global), which is why its cache barely grows with context.
How to run it
$ pip install mlx-lm $ mlx_lm.generate --model mlx-community/Olmo-3.1-32B-Instruct-4bit \ --max-tokens 512 --prompt "Hello"
Apple's own array framework. The fastest path on Apple Silicon. More on MLX.