can it run
No. 16GB of memory can't hold the 25GB working set this model needs. A smaller quant or a bigger machine is required.
weights 20GB + KV cache 1GB at 4k context vs ~16GB usable → no. 21.0 GB vs 16 GB usable — does not fit
computed roofline: 160.4 GB/s × 0.28–0.50 efficiency window / 20.0GB (Q4_K_M) — bands, never points.
| model file | 20GB (GGUF Q4_K_M) |
| minimum memory | 25GB working set |
| this machine | 16GB GDDR5 (2×8 GB), 160 GB/s |
| params | distill 32B |
| license | mit |
| price | $35 |
easy run: ollama pull deepseek-r1:32b
runs with: ollama · llama.cpp · koboldcpp · lm studio
→ Mac mini M5 Pro (64GB) — cheapest machine that runs it ($1669)
→ every model the Tesla M60 (16GB, used) can run — full list
→ tesla-p100-16gb — the next step up from this machine's GPU