gpu4.pics
data via LocalScore
NVIDIA RTX 6000 Ada Generation
48 GB VRAM NVIDIA
running Qwen2.5 14B Instruct· Q4_K_M· 14.8B
28.1
tokens / sec
560 · Strong #15 of 373
Generation
28.1tok/s
higher is better
Prompt / prefill
2,936tok/s
higher is better
Time to first token
471 ms
lower is better
Composite score
560
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 | 28.1 | 2,936 | 471 ms | — | — | — | 560 |