Intel Arc B580 12GBBudget pickC39.9 tok/s decode
~$249 MSRP
Can it run?
Tight fit
Using Q4_K_M in Ollama
Fit status
Tight fit
Decode
80.8 tok/s
TTFT
2396 ms
Safe context
19K
Memory
9.2 GB / 11.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs with offload | 72.9 tok/s | 3861 ms | 33K |
| Chat | B | Runs well | 72.9 tok/s | 1448 ms | 10K |
| Coding | C | Tight fit | 80.8 tok/s | 2396 ms | 19K |
| RAG | C | Runs with offload | 72.9 tok/s | 4826 ms | 33K |
| Reasoning | C | Tight fit | 72.9 tok/s | 3137 ms | 19K |
How GLM-4 9B (9B params) fits at each quantization level on RTX 2080 Ti 11GB (11.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 3.5 GB | Low | D35 |
Q3_K_S | 3 | 4.4 GB | Low | D37 |
NVFP4 | 4 |
ollama run glm-4-9bhuggingface-cli download glm-4-9bUpgrade options
Intel Arc B580 12GBBudget pick~$249 MSRP
RTX 3060 12GBBest value~$329 MSRP
RTX 3080 12GBBiggest leap~$799 MSRP
RTX 3080 Ti 12GBNVIDIA upgrade~$1,199 MSRP
| Medium |
| D38 |
Q4_K_M | 4 | 5.5 GB | Medium | D39 |
Q5_K_M | 5 | 6.5 GB | High | C41 |
Q6_KBest for your GPU | 6 | 7.4 GB | High | C43 |
Q8_0 | 8 | 9.6 GB | Very High | C44 |
F16 | 16 | 18.5 GB | Maximum | F0 |