TEST #493 RESULTS
04/07/2025 - 12:10 PM
ACCELERATOR
MODEL
31.6
tokens/s
generation
2.67
sec
time to first token
485
tokens/s
prompt
179
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 5800X3D 8-Core Processor (znver3)
RAM
31.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
580
tokens/s
37.1
tokens/s
1.79
sec
pp4096+tg256
450
tokens/s
18.9
tokens/s
9.16
sec
pp2048+tg256
528
tokens/s
28.1
tokens/s
3.91
sec
pp2048+tg768
528
tokens/s
26.1
tokens/s
3.92
sec
pp1024+tg1024
575
tokens/s
32.3
tokens/s
1.81
sec
pp1280+tg3072
550
tokens/s
24.3
tokens/s
2.36
sec
pp384+tg1152
594
tokens/s
36.8
tokens/s
670
ms
pp64+tg1024
222
tokens/s
41.7
tokens/s
309
ms
pp16+tg1536
342
tokens/s
39.2
tokens/s
66
ms