gpu4.pics
data via LocalScore
NVIDIA RTX 3090
llamafile 0.9.2
24 GB VRAM NVIDIA 350 W TDP
running Llama 3.2 1B Instruct· Q4_K_M· 1.5B
308
tokens / sec
3,294 · Excellent
Generation
308tok/s
higher is better
Prompt / prefill
12,344tok/s
higher is better
Time to first token
106 ms
lower is better
Composite score
3,294
Model size
1.5B
Runtime
llamafile
All benchmarks on this rig
| Model | Quant | Gen | Prompt | TTFT | Power | Perf/W | $/Mtok | Score |
|---|---|---|---|---|---|---|---|---|
| Llama 3.2 1B Instruct1.5B | Q4_K_M | 308 | 12,344 | 106 ms | — | — | — | 3,294 |