TEST #527 RESULTS
04/07/2025 - 8:48 PM
ACCELERATOR
MODEL
16.9
tokens/s
generation
521
ms
time to first token
2471
tokens/s
prompt
431
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 9800X3D 8-Core Processor
RAM
47.6GB
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
3399
tokens/s
20.7
tokens/s
350
ms
pp4096+tg256
2851
tokens/s
12.5
tokens/s
1.52
sec
pp2048+tg256
3130
tokens/s
17.3
tokens/s
709
ms
pp2048+tg768
3101
tokens/s
14.7
tokens/s
714
ms
pp1024+tg1024
3286
tokens/s
18.8
tokens/s
361
ms
pp1280+tg3072
2805
tokens/s
14.3
tokens/s
512
ms
pp384+tg1152
2830
tokens/s
18.4
tokens/s
209
ms
pp64+tg1024
666
tokens/s
17.4
tokens/s
160
ms
pp16+tg1536
171
tokens/s
17.9
tokens/s
150
ms