Yes — with 75.0 GB to spare
DeepSeek-R1-Distill-Qwen 32B at Q4_K_M fits your Ryzen AI Max+ 395 · 128 GB entirely in unified memory at 8K context, at an estimated 7.8 tokens per second. There is room for its full 128K window.
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 75.0 GB of 96.0 GB
Quantisation ladder
| Quant | Weights | Total @ 8K | Max context | Tok/s | Quality | Fit |
|---|---|---|---|---|---|---|
| Q8_0 | 32.5 GB | 35.1 GB | 128K | 4.4 | −0.1% ppl | Long context |
| Q6_K | 25.0 GB | 27.6 GB | 128K | 5.7 | −0.4% ppl | Long context |
| Q5_K_M | 21.7 GB | 24.3 GB | 128K | 6.6 | −0.8% ppl | Long context |
| Q4_K_M | 18.4 GB | 21.0 GB | 128K | 7.8 | −1.9% ppl | Recommended |
| Q3_K_M | 14.9 GB | 17.5 GB | 128K | 9.6 | −5.4% ppl | Long context |
| Q2_K | 12.8 GB | 15.4 GB | 128K | 11 | −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
$ ollama pull deepseek-r1:32b $ OLLAMA_CONTEXT_LENGTH=8192 \ ollama run deepseek-r1:32b
The default. One binary, a model registry, an OpenAI-compatible port. More on Ollama.
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 96.0 GB of 128 GB. Raising it with iogpu.wired_limit_mb is possible, and risky.
03There is room to go to the model's full 128K context on this card.