Can it run?
Runs well
Using Q4_K_M in Ollama
Fit status
Runs well
Decode
89.7 tok/s
TTFT
2158 ms
Safe context
27K
Memory
14.3 GB / 24.0 GB
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Agentic Coding | B | Runs well | 89.7 tok/s | 3139 ms | 47K |
| Chat | C | Runs well | 89.7 tok/s | 1177 ms | 15K |
| Coding | C | Runs well | 89.7 tok/s | 2158 ms | 27K |
| RAG | B | Runs well | 89.7 tok/s | 3924 ms | 47K |
| Reasoning | C | Runs well | 89.7 tok/s | 2551 ms | 27K |
How Qwen 2.5 14B (14B params) fits at each quantization level on RTX 4090 24GB (24.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 5.5 GB | Low | D34 |
Q3_K_S | 3 | 6.9 GB | Low | D35 |
NVFP4 | 4 | 7.8 GB | Medium | D36 |
Q4_K_M | 4 | 8.5 GB | Medium | D36 |
Q5_K_M | 5 | 10.1 GB | High | D38 |
Q6_K | 6 | 11.5 GB | High | D39 |
Q8_0Best for your GPU | 8 | 15.0 GB | Very High | C42 |
F16 | 16 | 28.7 GB | Maximum | F0 |
ollama run qwen-2.5-14bhuggingface-cli download qwen-2.5-14b