Can it run?
Runs well
Using Q4_K_M in Ollama
Fit status
Runs well
Decode
240.4 tok/s
TTFT
805 ms
Safe context
84K
Memory
17.7 GB / 92.2 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs well | 240.4 tok/s | 1171 ms | 166K |
| Chat | C | Runs well | 240.4 tok/s | 439 ms | 42K |
| Coding | C | Runs well | 240.4 tok/s | 805 ms | 84K |
| RAG | C | Runs well | 240.4 tok/s | 1464 ms | 166K |
| Reasoning | C | Runs well | 240.4 tok/s | 952 ms | 84K |
How Qwen2.5 3B Instruct (3B params) fits at each quantization level on Mac Studio M1 Ultra 128GB (92.2 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 1.2 GB | Low | D30 |
Q3_K_S | 3 | 1.5 GB | Low | D30 |
NVFP4 | 4 | 1.7 GB | Medium | D30 |
Q4_K_M | 4 | 1.8 GB | Medium | D30 |
Q5_K_M | 5 | 2.2 GB | High | D30 |
Q6_K | 6 | 2.5 GB | High | D30 |
Q8_0 | 8 | 3.2 GB | Very High | D30 |
F16Best for your GPU | 16 | 6.1 GB | Maximum | D31 |
