gpu4.pics
data via LocalScore
NVIDIA GeForce RTX 4060 Ti
llamafile 0.9.1
16 GB VRAM NVIDIA
running Qwen2.5 14B Instruct· Q4_K_M· 14.8B
26.9
tokens / sec
317 · Passable
Generation
26.9tok/s
higher is better
Prompt / prefill
1,266tok/s
higher is better
Time to first token
1.07 s
lower is better
Efficiency
0.2 tok/s/W
Electricity cost
$0.238/Mtok
Power draw
136 W
All benchmarks on this rig
| Model | Quant | Gen | Prompt | TTFT | Power | Perf/W | $/Mtok | Score |
|---|---|---|---|---|---|---|---|---|
| Qwen2.5 14B Instruct14.8B | Q4_K_M | 26.9 | 1,266 | 1.07 s | 136 W | 0.2 tok/s/W | $0.238/Mtok | 317 |