TEST #1827 RESULTS
11/19/2025 - 12:38 PM
ACCELERATOR
MODEL
26.6
tokens/s
generation
1.25
sec
time to first token
1018
tokens/s
prompt
278
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon Gold 5220R CPU @ 2.20GHz (cascadelake)
RAM
63.6GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1262
tokens/s
26.5
tokens/s
845
ms
pp4096+tg256
1016
tokens/s
25.5
tokens/s
4.07
sec
pp2048+tg256
1144
tokens/s
26.8
tokens/s
1.83
sec
pp2048+tg768
1125
tokens/s
26.4
tokens/s
1.86
sec
pp1024+tg1024
1184
tokens/s
27.0
tokens/s
904
ms
pp1280+tg3072
1130
tokens/s
25.9
tokens/s
1.17
sec
pp384+tg1152
1208
tokens/s
26.9
tokens/s
353
ms
pp64+tg1024
843
tokens/s
26.9
tokens/s
140
ms
pp16+tg1536
255
tokens/s
27.1
tokens/s
101
ms