TEST #83 RESULTS
03/28/2025 - 9:19 PM
ACCELERATOR
MODEL
31.4
tokens/s
generation
952
ms
time to first token
1408
tokens/s
prompt
359
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7352 24-Core Processor (znver2)
RAM
251.6GB
OS
Linux
Kernel Release
6.8.0-40-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #40~22.04.3-Ubuntu SMP PREEMPT_DYNAMIC Tue Jul 30 17:30:19 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1662
tokens/s
32.1
tokens/s
648
ms
pp4096+tg256
1275
tokens/s
29.1
tokens/s
3.25
sec
pp2048+tg256
1508
tokens/s
31.0
tokens/s
1.39
sec
pp2048+tg768
1510
tokens/s
30.7
tokens/s
1.39
sec
pp1024+tg1024
1661
tokens/s
31.6
tokens/s
649
ms
pp1280+tg3072
1614
tokens/s
30.4
tokens/s
825
ms
pp384+tg1152
1833
tokens/s
32.2
tokens/s
241
ms
pp64+tg1024
1284
tokens/s
32.7
tokens/s
102
ms
pp16+tg1536
320
tokens/s
32.5
tokens/s
80
ms