TEST #4092 RESULTS
08/27/2026 - 8:42 AM
ACCELERATOR
MODEL
49.5
tokens/s
generation
420
ms
time to first token
3060
tokens/s
prompt
711
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core Ultra 7 265K (arrowlake-s)
RAM
63.5GB
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
3762
tokens/s
51.3
tokens/s
292
ms
pp4096+tg256
2995
tokens/s
46.7
tokens/s
1.39
sec
pp2048+tg256
3481
tokens/s
48.0
tokens/s
610
ms
pp2048+tg768
3486
tokens/s
47.8
tokens/s
608
ms
pp1024+tg1024
3821
tokens/s
49.4
tokens/s
287
ms
pp1280+tg3072
3693
tokens/s
47.5
tokens/s
366
ms
pp384+tg1152
3950
tokens/s
50.8
tokens/s
131
ms
pp64+tg1024
1867
tokens/s
52.6
tokens/s
52
ms
pp16+tg1536
490
tokens/s
51.1
tokens/s
50
ms