TEST #1391 RESULTS
08/16/2025 - 3:07 PM
ACCELERATOR
MODEL
41.8
tokens/s
generation
627
ms
time to first token
2174
tokens/s
prompt
525
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i7-9700K CPU @ 3.60GHz (skylake)
RAM
62.7GB
OS
Linux
Kernel Release
6.8.0-71-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #71-Ubuntu SMP PREEMPT_DYNAMIC Tue Jul 22 16:52:38 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2473
tokens/s
40.7
tokens/s
439
ms
pp4096+tg256
1849
tokens/s
39.4
tokens/s
2.24
sec
pp2048+tg256
2411
tokens/s
42.8
tokens/s
873
ms
pp2048+tg768
2418
tokens/s
40.7
tokens/s
870
ms
pp1024+tg1024
2670
tokens/s
42.6
tokens/s
407
ms
pp1280+tg3072
2558
tokens/s
40.4
tokens/s
525
ms
pp384+tg1152
2959
tokens/s
42.7
tokens/s
155
ms
pp64+tg1024
1726
tokens/s
43.4
tokens/s
78
ms
pp16+tg1536
503
tokens/s
43.2
tokens/s
54
ms