TEST #1734 RESULTS
11/05/2025 - 6:01 AM
ACCELERATOR
MODEL
7.0
tokens/s
generation
17.28
sec
time to first token
70
tokens/s
prompt
30
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 9900X 12-Core Processor
RAM
31.1GB
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
86
tokens/s
7.9
tokens/s
11.99
sec
pp4096+tg256
86
tokens/s
7.4
tokens/s
47.92
sec
pp2048+tg256
98
tokens/s
7.9
tokens/s
20.93
sec
pp2048+tg768
99
tokens/s
6.9
tokens/s
20.91
sec
pp1024+tg1024
52
tokens/s
6.5
tokens/s
19.80
sec
pp1280+tg3072
52
tokens/s
6.3
tokens/s
24.79
sec
pp384+tg1152
53
tokens/s
6.6
tokens/s
7.33
sec
pp64+tg1024
53
tokens/s
6.7
tokens/s
1.35
sec
pp16+tg1536
47
tokens/s
6.7
tokens/s
487
ms