TEST #3498 RESULTS
05/10/2026 - 8:26 PM
ACCELERATOR
MODEL
30.7
tokens/s
generation
6.34
sec
time to first token
1169
tokens/s
prompt
178
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-1630 v3 @ 3.70GHz (haswell)
RAM
127.9GB
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
1976
tokens/s
33.5
tokens/s
568
ms
pp4096+tg256
95
tokens/s
23.6
tokens/s
43.15
sec
pp2048+tg256
1850
tokens/s
35.0
tokens/s
1.14
sec
pp2048+tg768
830
tokens/s
35.6
tokens/s
2.52
sec
pp1024+tg1024
2016
tokens/s
34.9
tokens/s
535
ms
pp1280+tg3072
148
tokens/s
12.4
tokens/s
8.68
sec
pp384+tg1152
2055
tokens/s
33.7
tokens/s
279
ms
pp64+tg1024
1208
tokens/s
33.8
tokens/s
82
ms
pp16+tg1536
338
tokens/s
33.9
tokens/s
81
ms