Intel Arc B580 12GBBudget pickC44.9 tok/s decode
~$249 MSRP
Can it run?
Runs with offload (needs ~0.1 GB host RAM)
Using Q4_K_M in Ollama
Fit status
Runs with offload (needs ~0.1 GB host RAM)
Decode
34.3 tok/s
TTFT
5637 ms
Safe context
16K
Memory
8.1 GB / 8.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Very compromised (needs ~0.7 GB host RAM) | 30.5 tok/s | 9243 ms | 27K |
| Chat | C | Runs with offload | 34.8 tok/s | 3033 ms | 8K |
| Coding | C | Runs with offload (needs ~0.1 GB host RAM) | 34.3 tok/s | 5637 ms | 16K |
| RAG | C | Very compromised (needs ~0.7 GB host RAM) | 30.5 tok/s | 11554 ms | 27K |
| Reasoning | C | Runs with offload (needs ~0.1 GB host RAM) | 34.3 tok/s | 6662 ms | 16K |
How Meta Llama 3 8B Instruct (8B params) fits at each quantization level on Radeon RX 7600M 8GB (8.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 3.1 GB | Low | D38 |
Q3_K_S | 3 | 3.9 GB | Low | D40 |
NVFP4 | 4 | 4.5 GB | Medium | C41 |
Q4_K_MBest for your GPU | 4 | 4.9 GB | Medium | C43 |
Q5_K_M | 5 | 5.8 GB | High | C44 |
Q6_K | 6 | 6.6 GB | High | C44 |
Q8_0 | 8 | 8.6 GB | Very High | F0 |
F16 | 16 | 16.4 GB | Maximum | F0 |
Upgrade options
Intel Arc B580 12GBBudget pick~$249 MSRP
RTX 3060 12GBBest value~$329 MSRP
~$449 MSRP
RTX 3080 12GBBiggest leap~$799 MSRP