Intel Arc B580 12GBBudget pickC51.3 tok/s decode
~$249 MSRP
Can it run?
Very compromised (needs ~0.7 GB host RAM)
Using Q4_K_M in Ollama
Fit status
Very compromised (needs ~0.7 GB host RAM)
Decode
22.6 tok/s
TTFT
8580 ms
Safe context
13K
Memory
7.2 GB / 6.0 GB
Offload
20%
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | F | Too heavy | 26.5 tok/s | 10615 ms | 23K |
| Chat | D | Very compromised (needs ~0.5 GB host RAM) | 23.4 tok/s | 4505 ms | 7K |
| Coding | D | Very compromised (needs ~0.7 GB host RAM) | 22.6 tok/s | 8580 ms | 13K |
| RAG | F | Too heavy | 26.5 tok/s | 13268 ms | 23K |
| Reasoning | D | Very compromised (needs ~0.7 GB host RAM) | 22.6 tok/s | 10140 ms | 13K |
How Llama 2 7B Chat (7B params) fits at each quantization level on GTX 1060 6GB (6.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 2.7 GB | Low | D39 |
Q3_K_SBest for your GPU | 3 | 3.4 GB | Low | C42 |
NVFP4 | 4 | 3.9 GB | Medium | C43 |
Q4_K_M | 4 | 4.3 GB | Medium | C44 |
Q5_K_M | 5 | 5.0 GB | High | C44 |
Q6_K | 6 | 5.7 GB | High | C44 |
Q8_0 | 8 | 7.5 GB | Very High | F0 |
F16 | 16 | 14.3 GB | Maximum | F0 |
Upgrade options
Intel Arc B580 12GBBudget pick~$249 MSRP
~$269 MSRP
RTX 3080 10GBNVIDIA upgrade~$699 MSRP
RTX 2080 Ti 11GBBiggest leap~$999 MSRP