TEST #1713 RESULTS
10/29/2025 - 4:33 PM
ACCELERATOR
MODEL
18.9
tokens/s
generation
2.56
sec
time to first token
490
tokens/s
prompt
153
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
13th Gen Intel Core i9-13900K (alderlake)
RAM
62.5GB
OS
Linux
Kernel Release
6.14.0-33-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #33~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Fri Sep 19 17:02:30 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
610
tokens/s
22.2
tokens/s
1.72
sec
pp4096+tg256
514
tokens/s
18.4
tokens/s
8.03
sec
pp2048+tg256
559
tokens/s
20.0
tokens/s
3.71
sec
pp2048+tg768
547
tokens/s
19.2
tokens/s
3.79
sec
pp1024+tg1024
561
tokens/s
18.9
tokens/s
1.87
sec
pp1280+tg3072
475
tokens/s
16.7
tokens/s
2.75
sec
pp384+tg1152
507
tokens/s
18.1
tokens/s
806
ms
pp64+tg1024
472
tokens/s
18.2
tokens/s
208
ms
pp16+tg1536
161
tokens/s
18.1
tokens/s
149
ms