TEST #1909 RESULTS
12/03/2025 - 9:14 PM
ACCELERATOR
MODEL
3.8
tokens/s
generation
8.84
sec
time to first token
289
tokens/s
prompt
50
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i9-14900HX (alderlake)
RAM
95.7GB
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
127
tokens/s
3.8
tokens/s
8.33
sec
pp4096+tg256
78
tokens/s
1.9
tokens/s
53.16
sec
pp2048+tg256
456
tokens/s
4.2
tokens/s
4.74
sec
pp2048+tg768
437
tokens/s
3.9
tokens/s
4.93
sec
pp1024+tg1024
460
tokens/s
4.1
tokens/s
2.48
sec
pp1280+tg3072
394
tokens/s
3.5
tokens/s
3.54
sec
pp384+tg1152
381
tokens/s
4.3
tokens/s
1.29
sec
pp64+tg1024
221
tokens/s
4.5
tokens/s
559
ms
pp16+tg1536
47
tokens/s
4.3
tokens/s
572
ms