TEST #3506 RESULTS
05/11/2026 - 8:32 PM
ACCELERATOR
MODEL
28.9
tokens/s
generation
1.21
sec
time to first token
907
tokens/s
prompt
279
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7251 8-Core Processor
RAM
188.8GB
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
1379
tokens/s
50.4
tokens/s
762
ms
pp4096+tg256
1282
tokens/s
40.1
tokens/s
3.22
sec
pp2048+tg256
1316
tokens/s
35.4
tokens/s
1.58
sec
pp2048+tg768
1324
tokens/s
28.6
tokens/s
1.57
sec
pp1024+tg1024
918
tokens/s
24.1
tokens/s
1.14
sec
pp1280+tg3072
838
tokens/s
20.3
tokens/s
1.56
sec
pp384+tg1152
702
tokens/s
19.9
tokens/s
579
ms
pp64+tg1024
180
tokens/s
20.9
tokens/s
408
ms
pp16+tg1536
225
tokens/s
20.7
tokens/s
90
ms