TEST #1817 RESULTS
11/16/2025 - 3:55 PM
ACCELERATOR
MODEL
18.2
tokens/s
generation
1.26
sec
time to first token
1078
tokens/s
prompt
250
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
RAM
119.7GB
OS
Linux
Kernel Release
6.11.0-1016-nvidia
Architecture
aarch64
Version
Cosmopolitan 3.9.7 MODE=aarch64; #16-Ubuntu SMP PREEMPT_DYNAMIC Sun Sep 21 16:52:46 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1328
tokens/s
18.5
tokens/s
823
ms
pp4096+tg256
955
tokens/s
17.3
tokens/s
4.35
sec
pp2048+tg256
1174
tokens/s
17.6
tokens/s
1.80
sec
pp2048+tg768
1130
tokens/s
17.6
tokens/s
1.87
sec
pp1024+tg1024
1268
tokens/s
18.3
tokens/s
862
ms
pp1280+tg3072
1310
tokens/s
17.4
tokens/s
1.03
sec
pp384+tg1152
1508
tokens/s
19.0
tokens/s
309
ms
pp64+tg1024
806
tokens/s
19.3
tokens/s
163
ms
pp16+tg1536
225
tokens/s
18.8
tokens/s
122
ms