Can it run?
Runs well
Using Q4_K_M in Ollama
Fit status
Runs well
Decode
102.3 tok/s
TTFT
1893 ms
Safe context
34K
Memory
9.3 GB / 20.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs well | 102.3 tok/s | 2753 ms | 60K |
| Chat | C | Runs well | 102.3 tok/s | 1032 ms | 18K |
| Coding | C | Runs well | 102.3 tok/s | 1893 ms | 34K |
| RAG | C | Runs well | 102.3 tok/s | 3441 ms | 60K |
| Reasoning | C | Runs well | 102.3 tok/s | 2237 ms | 34K |
How llava llama 3 8b v1 1 (8B params) fits at each quantization level on RTX A4500 20GB (20.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 3.1 GB | Low | D33 |
Q3_K_S | 3 | 3.9 GB | Low | D34 |
NVFP4 | 4 | 4.5 GB | Medium | D34 |
Q4_K_M | 4 | 4.9 GB | Medium | D35 |
Q5_K_M | 5 | 5.8 GB | High | D36 |
Q6_K | 6 | 6.6 GB | High | D37 |
Q8_0Best for your GPU | 8 | 8.6 GB | Very High | D39 |
F16 | 16 | 16.4 GB | Maximum | C44 |
