Intel Arc B580 12GBBudget pickC89.7 tok/s decode
~$249 MSRP
Can it run?
Too heavy
Using Q4_K_M in Ollama
Fit status
Too heavy
Decode
26.2 tok/s
TTFT
7381 ms
Safe context
13K
Memory
4.8 GB / 4.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | F | Too heavy | 26.2 tok/s | 10736 ms | 24K |
| Chat | F | Too heavy | 26.2 tok/s | 4026 ms | 7K |
| Coding | F | Too heavy | 26.2 tok/s | 7381 ms | 13K |
| RAG | F | Too heavy | 26.2 tok/s | 13420 ms | 24K |
| Reasoning | F | Too heavy | 26.2 tok/s | 8723 ms | 13K |
How gemma 3 4b it (4B params) fits at each quantization level on GTX 1650 4GB (4.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_KBest for your GPU | 2 | 1.6 GB | Low | D38 |
Q3_K_S | 3 | 2.0 GB | Low | C40 |
NVFP4 | 4 | 2.2 GB | Medium | C41 |
Q4_K_M | 4 | 2.4 GB | Medium | C42 |
Q5_K_M | 5 | 2.9 GB | High | C44 |
Q6_K | 6 | 3.3 GB | High | C44 |
Q8_0 | 8 | 4.3 GB | Very High | F0 |
F16 | 16 | 8.2 GB | Maximum | F0 |
Upgrade options
Intel Arc B580 12GBBudget pick~$249 MSRP
~$269 MSRP
RTX 2060 Super 8GBNVIDIA upgrade~$399 MSRP
Intel Arc A580 8GBBiggest leap