← CanIRunLocal · All hardware · All models
Yes. Ministral 3 14B fits in fast memory on Mac Studio M4 Max (32c GPU) 36GB at Q8_0 (16.6 GB needed, 27.0 GB usable), about 22 tok/s (estimate).
Hardware: Mac Studio M4 Max (32c GPU) 36GB · Model: Ministral 3 14B
Mac Studio M4 Max (32c GPU) 36GB has 36 GB of unified memory and 410 GB/s of memory bandwidth. M4 Max (32c GPU) 36GB has the same memory and bandwidth, so the fit numbers match; this page is Mac Studio M4 Max (32c GPU) 36GB. The checker leaves macOS a share of unified memory, so Ollama can use 27.0 GB. Ministral 3 14B (13.9B, Mistral) at Q8_0 is a 15 GB download and needs 16.6 GB with headroom for about 8k tokens.
| Best quant | Q8_0 |
|---|---|
| Download size | 15 GB |
| Memory needed | 16.6 GB |
| Usable fast memory | 27.0 GB |
| Fit | fits in fast memory |
| Estimated speed | ~22 tok/s (estimate) |
| Command | ollama run ministral-3:14b-instruct-2512-q8_0 |
Estimates, not benchmarks. Model sizes are ollama.com download sizes; memory needed adds headroom for ~8k tokens of context; speeds are rough figures from memory bandwidth. Data updated 2026-10-23.
A project by OnDevice Labs · llms.txt