TEST #1112 RESULTS
06/25/2025 - 11:12 AM
ACCELERATOR
MODEL
40.9
tokens/s
generation
680
ms
time to first token
1852
tokens/s
prompt
481
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon W-2235 CPU @ 3.80GHz (cascadelake)
RAM
191.6GB
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
2273
tokens/s
39.5
tokens/s
492
ms
pp4096+tg256
1913
tokens/s
39.5
tokens/s
2.17
sec
pp2048+tg256
2134
tokens/s
41.5
tokens/s
998
ms
pp2048+tg768
2124
tokens/s
41.4
tokens/s
1.00
sec
pp1024+tg1024
2234
tokens/s
41.3
tokens/s
483
ms
pp1280+tg3072
2169
tokens/s
40.0
tokens/s
616
ms
pp384+tg1152
2188
tokens/s
41.6
tokens/s
215
ms
pp64+tg1024
1321
tokens/s
41.6
tokens/s
71
ms
pp16+tg1536
311
tokens/s
42.0
tokens/s
73
ms