TEST #3515 RESULTS
05/11/2026 - 10:45 PM
ACCELERATOR
MODEL
32.8
tokens/s
generation
1.26
sec
time to first token
1028
tokens/s
prompt
299
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7251 8-Core Processor
RAM
220.3GB
OS
Linux
Kernel Release
6.17.0-23-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #23~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Tue Apr 14 16:11:48 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1239
tokens/s
34.5
tokens/s
856
ms
pp4096+tg256
1000
tokens/s
30.0
tokens/s
4.13
sec
pp2048+tg256
1112
tokens/s
32.5
tokens/s
1.87
sec
pp2048+tg768
1098
tokens/s
31.9
tokens/s
1.90
sec
pp1024+tg1024
1162
tokens/s
33.0
tokens/s
912
ms
pp1280+tg3072
1118
tokens/s
31.0
tokens/s
1.17
sec
pp384+tg1152
1210
tokens/s
33.7
tokens/s
346
ms
pp64+tg1024
992
tokens/s
34.3
tokens/s
112
ms
pp16+tg1536
322
tokens/s
34.0
tokens/s
78
ms