TEST #3480 RESULTS
05/09/2026 - 1:49 PM
ACCELERATOR
MODEL
44.9
tokens/s
generation
708
ms
time to first token
1762
tokens/s
prompt
482
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 5900X 12-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
2117
tokens/s
48.2
tokens/s
505
ms
pp4096+tg256
1787
tokens/s
43.4
tokens/s
2.32
sec
pp2048+tg256
2050
tokens/s
47.8
tokens/s
1.02
sec
pp2048+tg768
2026
tokens/s
46.1
tokens/s
1.03
sec
pp1024+tg1024
2213
tokens/s
48.1
tokens/s
484
ms
pp1280+tg3072
2139
tokens/s
45.0
tokens/s
619
ms
pp384+tg1152
2283
tokens/s
41.1
tokens/s
205
ms
pp64+tg1024
829
tokens/s
39.7
tokens/s
132
ms
pp16+tg1536
411
tokens/s
44.7
tokens/s
60
ms