TEST #10 RESULTS
03/19/2025 - 10:30 PM
ACCELERATOR
MODEL
26.9
tokens/s
generation
1.07
sec
time to first token
1266
tokens/s
prompt
317
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7352 24-Core Processor (znver2)
RAM
125.64GB
OS
Linux
Kernel Release
6.8.0-52-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #53~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Wed Jan 15 19:18:46 UTC 2
RUNTIME
Name
llamafile
Version
0.9.1
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1358
tokens/s
27.4
tokens/s
791
ms
pp4096+tg256
1115
tokens/s
24.2
tokens/s
3.71
sec
pp2048+tg256
1347
tokens/s
26.6
tokens/s
1.56
sec
pp2048+tg768
1369
tokens/s
26.4
tokens/s
1.53
sec
pp1024+tg1024
1522
tokens/s
27.3
tokens/s
709
ms
pp1280+tg3072
1486
tokens/s
26.1
tokens/s
898
ms
pp384+tg1152
1670
tokens/s
27.9
tokens/s
265
ms
pp64+tg1024
1191
tokens/s
28.3
tokens/s
110
ms
pp16+tg1536
338
tokens/s
28.0
tokens/s
82
ms