TEST #3387 RESULTS
04/26/2026 - 3:10 PM
ACCELERATOR
MODEL
19.2
tokens/s
generation
6.43
sec
time to first token
181
tokens/s
prompt
81
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-2680 v4 @ 2.40GHz (broadwell)
RAM
31.2GB
OS
Linux
Kernel Release
6.17.13-3-pve
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC PMX 6.17.13-3 (2026-03-31T21:50Z)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
221
tokens/s
20.1
tokens/s
4.68
sec
pp4096+tg256
206
tokens/s
16.7
tokens/s
19.91
sec
pp2048+tg256
212
tokens/s
18.7
tokens/s
9.69
sec
pp2048+tg768
211
tokens/s
18.4
tokens/s
9.74
sec
pp1024+tg1024
216
tokens/s
19.4
tokens/s
4.78
sec
pp1280+tg3072
213
tokens/s
18.0
tokens/s
6.06
sec
pp384+tg1152
223
tokens/s
20.1
tokens/s
1.77
sec
pp64+tg1024
91
tokens/s
20.6
tokens/s
751
ms
pp16+tg1536
34
tokens/s
20.3
tokens/s
515
ms