Can it run?
Runs well
Using Q4_K_M in Ollama
Fit status
Runs well
Decode
233.3 tok/s
TTFT
830 ms
Safe context
26K
Memory
3.6 GB / 6.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | C | Runs well | 178.2 tok/s | 1580 ms | 33K |
| Chat | C | Runs well | 178.2 tok/s | 593 ms | 13K |
| Coding | C | Runs well | 233.3 tok/s | 830 ms | 26K |
| RAG | C | Runs well | 178.2 tok/s | 1975 ms | 33K |
| Reasoning | C | Runs well | 178.2 tok/s | 1284 ms | 26K |
How Qwen 3 1.7B (1.7000000476837158B params) fits at each quantization level on GTX 1660 Super 6GB (6.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 0.7 GB | Low | D30 |
Q3_K_S | 3 | 0.8 GB | Low | D31 |
NVFP4 | 4 |
ollama run qwen-3-1.7bhuggingface-cli download qwen-3-1.7b1.0 GB |
| Medium |
| D32 |
Q4_K_M | 4 | 1.0 GB | Medium | D32 |
Q5_K_M | 5 | 1.2 GB | High | D32 |
Q6_K | 6 | 1.4 GB | High | D33 |
Q8_0 | 8 | 1.8 GB | Very High | D34 |
F16Best for your GPU | 16 | 3.5 GB | Maximum | C40 |