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
13.6 tok/s
TTFT
14194 ms
Safe context
9K
Memory
20.8 GB / 12.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | F | Too heavy | 13.6 tok/s | 20646 ms | 16K |
| Chat | F | Too heavy | 13.6 tok/s | 7742 ms | 5K |
| Coding | F | Too heavy | 13.6 tok/s | 14194 ms | 9K |
| RAG | F | Too heavy | 13.6 tok/s | 25808 ms | 16K |
| Reasoning | F | Too heavy | 13.6 tok/s | 16775 ms | 9K |
How Devstral Small 2 24B Instruct (24B params) fits at each quantization level on RX 6700 XT 12GB (12.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 9.4 GB | Low | C44 |
Q3_K_S | 3 | 11.8 GB | Low | C44 |
NVFP4 | 4 | 13.4 GB | Medium | F0 |
Q4_K_M | 4 | 14.6 GB | Medium | F0 |
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 |
Upgrade options
~$799 MSRP
~$999 MSRP
RTX 5090 32GBBiggest leap~$1,999 MSRP