TEST #1024 RESULTS
06/09/2025 - 5:51 PM
ACCELERATOR
MODEL
28.4
tokens/s
generation
1.55
sec
time to first token
821
tokens/s
prompt
247
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon W-11855M CPU @ 3.20GHz (tigerlake)
RAM
31.1GB
OS
Linux
Kernel Release
6.12.28-1-MANJARO
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Fri, 09 May 2025 10:53:27 +0000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1023
tokens/s
29.7
tokens/s
1.04
sec
pp4096+tg256
826
tokens/s
25.9
tokens/s
5.00
sec
pp2048+tg256
899
tokens/s
27.9
tokens/s
2.32
sec
pp2048+tg768
897
tokens/s
27.7
tokens/s
2.32
sec
pp1024+tg1024
899
tokens/s
28.8
tokens/s
1.17
sec
pp1280+tg3072
909
tokens/s
27.2
tokens/s
1.44
sec
pp384+tg1152
932
tokens/s
29.4
tokens/s
452
ms
pp64+tg1024
725
tokens/s
29.8
tokens/s
140
ms
pp16+tg1536
280
tokens/s
29.5
tokens/s
93
ms