Intel Arc B580 12GBBudget pickC51.3 tok/s decode
~$249 MSRP
Can it run?
Runs well
Using Q4_K_M in Ollama
Fit status
Runs well
Decode
30.4 tok/s
TTFT
6359 ms
Safe context
22K
Memory
8.3 GB / 11.5 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs well | 32.8 tok/s | 8589 ms | 39K |
| Chat | C | Runs well | 32.8 tok/s | 3221 ms | 12K |
| Coding | C | Runs well | 30.4 tok/s | 6359 ms | 22K |
| RAG | C | Runs well | 32.8 tok/s | 10736 ms | 39K |
| Reasoning | C | Runs well | 32.8 tok/s | 6978 ms | 22K |
How Llama 2 7B Chat (7B params) fits at each quantization level on MacBook Pro M2 Pro 16GB (11.5 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 2.7 GB | Low | D35 |
Q3_K_S | 3 | 3.4 GB | Low | D36 |
NVFP4 | 4 |
Upgrade options
Intel Arc B580 12GBBudget pick~$249 MSRP
RTX 3060 12GBBest value~$329 MSRP
RTX 3080 12GBBiggest leap~$799 MSRP
3.9 GB |
| Medium |
| D37 |
Q4_K_M | 4 | 4.3 GB | Medium | D37 |
Q5_K_M | 5 | 5.0 GB | High | D39 |
Q6_K | 6 | 5.7 GB | High | C40 |
Q8_0Best for your GPU | 8 | 7.5 GB | Very High | C43 |
F16 | 16 | 14.3 GB | Maximum | F0 |