gpu4.pics
data via LocalScore
NVIDIA RTX 5090
32 GB VRAM NVIDIA 575 W TDP
running Qwen2.5 14B Instruct· Q4_K_M· 14.8B
45.5
tokens / sec
708 · Strong #10 of 373
Generation
45.5tok/s
higher is better
Prompt / prefill
3,678tok/s
higher is better
Time to first token
536 ms
lower is better
Composite score
708
Model size
14.8B
Runtime
—
All benchmarks on this rig
| Model | Quant | Gen | Prompt | TTFT | Power | Perf/W | $/Mtok | Score |
|---|---|---|---|---|---|---|---|---|
| Qwen2.5 14B Instruct14.8B | Q4_K_M | 45.5 | 3,678 | 536 ms | — | — | — | 708 |