Can it run?
Runs well
Using Q4_K_M in Ollama
Fit status
Runs well
Decode
59.8 tok/s
TTFT
3237 ms
Safe context
27K
Memory
7.0 GB / 12.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs well | 59.8 tok/s | 4709 ms | 48K |
| Chat | C | Runs well | 59.8 tok/s | 1766 ms | 14K |
| Coding | C | Runs well | 59.8 tok/s | 3237 ms | 27K |
| RAG | C | Runs well | 59.8 tok/s | 5886 ms | 48K |
| Reasoning | C | Runs well | 59.8 tok/s | 3826 ms | 27K |
How stablelm 2 1 6b chat imatrix (6B params) fits at each quantization level on Intel Arc B580 12GB (12.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 2.3 GB | Low | D33 |
Q3_K_S | 3 | 2.9 GB | Low | D34 |
NVFP4 | 4 | 3.4 GB | Medium | D35 |
Q4_K_M | 4 | 3.7 GB | Medium | D36 |
Q5_K_M | 5 | 4.3 GB | High | D37 |
Q6_K | 6 | 4.9 GB | High | D38 |
Q8_0Best for your GPU | 8 | 6.4 GB | Very High | C40 |
F16 | 16 | 12.3 GB | Maximum | F0 |
