TEST #939 RESULTS
05/24/2025 - 1:45 AM
ACCELERATOR
MODEL
81.1
tokens/s
generation
245
ms
time to first token
5126
tokens/s
prompt
1193
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen Threadripper 3970X 32-Core Processor (znver2)
RAM
125.6GB
OS
Linux
Kernel Release
6.8.0-60-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #63-Ubuntu SMP PREEMPT_DYNAMIC Tue Apr 15 19:04:15 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
5625
tokens/s
82.7
tokens/s
194
ms
pp4096+tg256
5200
tokens/s
72.0
tokens/s
812
ms
pp2048+tg256
6229
tokens/s
79.9
tokens/s
341
ms
pp2048+tg768
6285
tokens/s
78.9
tokens/s
338
ms
pp1024+tg1024
6703
tokens/s
82.6
tokens/s
165
ms
pp1280+tg3072
6433
tokens/s
77.4
tokens/s
212
ms
pp384+tg1152
5944
tokens/s
84.6
tokens/s
76
ms
pp64+tg1024
2734
tokens/s
86.7
tokens/s
35
ms
pp16+tg1536
982
tokens/s
84.7
tokens/s
28
ms