TEST #1699 RESULTS
10/25/2025 - 3:58 AM
ACCELERATOR
MODEL
19.4
tokens/s
generation
3.39
sec
time to first token
381
tokens/s
prompt
130
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i7-3770 CPU @ 3.40GHz (ivybridge)
RAM
15.5GB
OS
Linux
Kernel Release
6.8.0-86-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #87-Ubuntu SMP PREEMPT_DYNAMIC Mon Sep 22 18:03:36 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
400
tokens/s
19.8
tokens/s
2.61
sec
pp4096+tg256
373
tokens/s
15.3
tokens/s
11.06
sec
pp2048+tg256
414
tokens/s
18.8
tokens/s
5.00
sec
pp2048+tg768
406
tokens/s
18.3
tokens/s
5.09
sec
pp1024+tg1024
436
tokens/s
20.0
tokens/s
2.39
sec
pp1280+tg3072
425
tokens/s
17.4
tokens/s
3.06
sec
pp384+tg1152
443
tokens/s
21.2
tokens/s
911
ms
pp64+tg1024
353
tokens/s
22.3
tokens/s
226
ms
pp16+tg1536
178
tokens/s
21.8
tokens/s
136
ms