TEST #1413 RESULTS
08/23/2025 - 5:37 AM
ACCELERATOR
MODEL
65.5
tokens/s
generation
555
ms
time to first token
2320
tokens/s
prompt
650
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 3900X 12-Core Processor (znver2)
RAM
125.7GB
OS
Linux
Kernel Release
6.14.0-28-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #28~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Fri Jul 25 10:47:01 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2753
tokens/s
68.2
tokens/s
387
ms
pp4096+tg256
2246
tokens/s
58.7
tokens/s
1.84
sec
pp2048+tg256
2560
tokens/s
64.4
tokens/s
816
ms
pp2048+tg768
2559
tokens/s
63.7
tokens/s
817
ms
pp1024+tg1024
2755
tokens/s
66.6
tokens/s
386
ms
pp1280+tg3072
2666
tokens/s
62.4
tokens/s
495
ms
pp384+tg1152
2842
tokens/s
68.0
tokens/s
161
ms
pp64+tg1024
1952
tokens/s
69.2
tokens/s
47
ms
pp16+tg1536
551
tokens/s
68.5
tokens/s
43
ms