TEST #1802 RESULTS
11/15/2025 - 9:40 PM
ACCELERATOR
MODEL
36.0
tokens/s
generation
788
ms
time to first token
1681
tokens/s
prompt
425
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
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2008
tokens/s
34.9
tokens/s
538
ms
pp4096+tg256
1555
tokens/s
34.4
tokens/s
2.66
sec
pp2048+tg256
1800
tokens/s
36.2
tokens/s
1.17
sec
pp2048+tg768
1836
tokens/s
36.4
tokens/s
1.14
sec
pp1024+tg1024
2020
tokens/s
36.7
tokens/s
534
ms
pp1280+tg3072
1961
tokens/s
35.9
tokens/s
681
ms
pp384+tg1152
2168
tokens/s
36.6
tokens/s
204
ms
pp64+tg1024
1359
tokens/s
36.7
tokens/s
102
ms
pp16+tg1536
419
tokens/s
36.5
tokens/s
64
ms