Intel Arc B580 12GBBudget pickC119.6 tok/s decode
~$249 MSRP
Can it run?
Runs well
Using Q4_K_M in Ollama
Fit status
Runs well
Decode
47.2 tok/s
TTFT
4101 ms
Safe context
33K
Memory
5.6 GB / 11.5 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs well | 47.2 tok/s | 5964 ms | 65K |
| Chat | C | Runs well | 47.2 tok/s | 2237 ms | 17K |
| Coding | C | Runs well | 47.2 tok/s | 4101 ms | 33K |
| RAG | C | Runs well | 47.2 tok/s | 7456 ms | 65K |
| Reasoning | C | Runs well | 47.2 tok/s | 4846 ms | 33K |
How Qwen2.5 3B Instruct (3B params) fits at each quantization level on MacBook Pro M4 16GB (11.5 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 1.2 GB | Low | D32 |
Q3_K_S | 3 | 1.5 GB | Low | D32 |
NVFP4 | 4 |
Upgrade options
Intel Arc B580 12GBBudget pick~$249 MSRP
RTX 3060 12GBBest value~$329 MSRP
RTX A2000 12GBBiggest leap
1.7 GB |
| Medium |
| D33 |
Q4_K_M | 4 | 1.8 GB | Medium | D33 |
Q5_K_M | 5 | 2.2 GB | High | D34 |
Q6_K | 6 | 2.5 GB | High | D34 |
Q8_0 | 8 | 3.2 GB | Very High | D35 |
F16Best for your GPU | 16 | 6.1 GB | Maximum | C41 |