TEST #3349 RESULTS
04/19/2026 - 5:03 AM
ACCELERATOR
MODEL
32.4
tokens/s
generation
2.62
sec
time to first token
489
tokens/s
prompt
182
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 5950X 16-Core Processor (znver3)
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
592
tokens/s
37.7
tokens/s
1.76
sec
pp4096+tg256
462
tokens/s
19.3
tokens/s
8.92
sec
pp2048+tg256
537
tokens/s
28.3
tokens/s
3.85
sec
pp2048+tg768
536
tokens/s
26.7
tokens/s
3.86
sec
pp1024+tg1024
580
tokens/s
33.0
tokens/s
1.79
sec
pp1280+tg3072
558
tokens/s
24.9
tokens/s
2.32
sec
pp384+tg1152
585
tokens/s
38.0
tokens/s
679
ms
pp64+tg1024
211
tokens/s
43.2
tokens/s
323
ms
pp16+tg1536
345
tokens/s
40.4
tokens/s
66
ms