TEST #1103 RESULTS
06/24/2025 - 7:52 PM
ACCELERATOR
MODEL
34.3
tokens/s
generation
1.01
sec
time to first token
1320
tokens/s
prompt
355
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-1620 v3 @ 3.50GHz (haswell)
RAM
15.5GB
OS
Linux
Kernel Release
6.1.0-37-amd64
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Debian 6.1.140-1 (2025-05-22)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1133
tokens/s
34.8
tokens/s
932
ms
pp4096+tg256
1224
tokens/s
31.1
tokens/s
3.38
sec
pp2048+tg256
1458
tokens/s
33.9
tokens/s
1.43
sec
pp2048+tg768
1464
tokens/s
33.6
tokens/s
1.43
sec
pp1024+tg1024
1607
tokens/s
34.8
tokens/s
666
ms
pp1280+tg3072
1556
tokens/s
33.0
tokens/s
852
ms
pp384+tg1152
1776
tokens/s
35.7
tokens/s
244
ms
pp64+tg1024
1215
tokens/s
36.3
tokens/s
111
ms
pp16+tg1536
446
tokens/s
35.7
tokens/s
63
ms