TEST #3065 RESULTS
03/12/2026 - 1:35 AM
ACCELERATOR
MODEL
16.6
tokens/s
generation
1.44
sec
time to first token
953
tokens/s
prompt
222
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 9950X 16-Core Processor
RAM
123.4GB
OS
Linux
Kernel Release
6.17.9-1-pve
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC PMX 6.17.9-1 (2026-01-12T16:25Z)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1167
tokens/s
19.1
tokens/s
930
ms
pp4096+tg256
801
tokens/s
11.7
tokens/s
5.20
sec
pp2048+tg256
1023
tokens/s
14.7
tokens/s
2.07
sec
pp2048+tg768
1023
tokens/s
14.6
tokens/s
2.07
sec
pp1024+tg1024
1207
tokens/s
16.1
tokens/s
902
ms
pp1280+tg3072
1130
tokens/s
14.0
tokens/s
1.20
sec
pp384+tg1152
1322
tokens/s
18.7
tokens/s
337
ms
pp64+tg1024
663
tokens/s
21.2
tokens/s
139
ms
pp16+tg1536
240
tokens/s
19.5
tokens/s
109
ms