Yes, just — 3.0 GB spare
DeepSeek-R1-Distill-Qwen 32B at Q4_K_M fits your M1 Pro · 32 GB entirely in unified memory at 8K context, at an estimated 6.1 tokens per second. Past 19K the KV cache pushes it over — quantise the cache to q8_0, or step down a quantisation, to go longer.
Fully in unified memory
8K context
Q4_K_M · 18.4 GB
MIT
Released Jan 2025
MIT-licensed and close to the 70B distill on maths. A 24 GB card handles it at Q4.
The VRAM budget
weights 18.4 GB
Weights 18.4 GB
KV cache @ 8K 2.00 GB
Runtime overhead 0.6 GB
Free 3.0 GB of 24.0 GB
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| Q8_0 | 32.5 GB | 35.1 GB | — | ~3.4 | −0.1% ppl | 11.1 GB over |
| Q6_K | 25.0 GB | 27.6 GB | — | ~4.5 | −0.4% ppl | 3.6 GB over |
| Q5_K_M | 21.7 GB | 24.3 GB | 6K | ~5.2 | −0.8% ppl | 0.3 GB over |
| Q4_K_M | 18.4 GB | 21.0 GB | 19K | 6.1 | −1.9% ppl | Recommended |
| Q3_K_M | 14.9 GB | 17.5 GB | 33K | 7.5 | −5.4% ppl | Long context |
| Q2_K | 12.8 GB | 15.4 GB | 42K | 8.7 | −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.
How to run it
$ pip install mlx-lm $ mlx_lm.generate --model mlx-community/DeepSeek-R1-Distill-Qwen-32B-4bit \ --max-tokens 512 --prompt "Hello"
Apple's own array framework. The fastest path on Apple Silicon. More on MLX.
01Download is 18.4 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 24.0 GB of 32 GB. Raising it with iogpu.wired_limit_mb is possible, and risky.
03Only 3.0 GB is spare, so a long system prompt can still push it over. Its real ceiling here is 19K context.