TEST #3508 RESULTS
05/11/2026 - 9:44 PM
ACCELERATOR
MODEL
32.8
tokens/s
generation
1.24
sec
time to first token
1029
tokens/s
prompt
301
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
1183
tokens/s
33.1
tokens/s
894
ms
pp4096+tg256
1017
tokens/s
30.3
tokens/s
4.06
sec
pp2048+tg256
1154
tokens/s
32.7
tokens/s
1.81
sec
pp2048+tg768
1147
tokens/s
32.2
tokens/s
1.82
sec
pp1024+tg1024
1201
tokens/s
33.2
tokens/s
883
ms
pp1280+tg3072
1144
tokens/s
31.4
tokens/s
1.15
sec
pp384+tg1152
1208
tokens/s
34.0
tokens/s
347
ms
pp64+tg1024
892
tokens/s
34.5
tokens/s
124
ms
pp16+tg1536
318
tokens/s
34.2
tokens/s
87
ms