TEST #3521 RESULTS
05/12/2026 - 10:33 PM
ACCELERATOR
MODEL
32.7
tokens/s
generation
1.27
sec
time to first token
1002
tokens/s
prompt
296
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
1161
tokens/s
33.5
tokens/s
911
ms
pp4096+tg256
1004
tokens/s
30.0
tokens/s
4.11
sec
pp2048+tg256
1120
tokens/s
32.3
tokens/s
1.86
sec
pp2048+tg768
1109
tokens/s
31.9
tokens/s
1.88
sec
pp1024+tg1024
1172
tokens/s
33.0
tokens/s
904
ms
pp1280+tg3072
1126
tokens/s
31.1
tokens/s
1.17
sec
pp384+tg1152
1194
tokens/s
33.8
tokens/s
350
ms
pp64+tg1024
863
tokens/s
34.3
tokens/s
126
ms
pp16+tg1536
269
tokens/s
34.0
tokens/s
88
ms