MacBook Pro M4 32GBBest value
C5.9 tok/s decode
~$799 MSRP
Can it run?
Too heavy
Using Q4_K_M in Ollama
Fit status
Too heavy
Decode
12.1 tok/s
TTFT
15972 ms
Safe context
12K
Memory
22.2 GB / 17.3 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | F | Too heavy | 12.1 tok/s | 23233 ms | 21K |
| Chat | D | Very compromised | 11.6 tok/s | 9120 ms | 7K |
| Coding | F | Too heavy | 12.1 tok/s | 15972 ms | 12K |
| RAG | F | Too heavy | 12.1 tok/s | 29041 ms | 21K |
| Reasoning | F | Too heavy | 12.1 tok/s | 18876 ms | 12K |
How Devstral Small 2 24B Instruct (24B params) fits at each quantization level on MacBook Pro M4 Pro 24GB (17.3 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 9.4 GB | Low | C41 |
Q3_K_SBest for your GPU | 3 | 11.8 GB | Low | C44 |
Upgrade options
~$799 MSRP
~$999 MSRP
RTX 5090 32GBBiggest leapNVFP4 |
| 4 |
13.4 GB |
| Medium |
| C44 |
Q4_K_M | 4 | 14.6 GB | Medium | C44 |
Q5_K_M | 5 | 17.3 GB | High | F0 |
Q6_K | 6 | 19.7 GB | High | F0 |
Q8_0 | 8 | 25.7 GB | Very High | F0 |
F16 | 16 | 49.2 GB | Maximum | F0 |
~$1,999 MSRP
~$2,499 MSRP