TEST #754 RESULTS
04/22/2025 - 12:08 AM
ACCELERATOR
MODEL
23.7
tokens/s
generation
558
ms
time to first token
2291
tokens/s
prompt
460
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 9950X 16-Core Processor
RAM
60.4GB
OS
Linux
Kernel Release
6.14.0-061400-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #202503241442 SMP PREEMPT_DYNAMIC Mon Mar 24 15:06:01 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2924
tokens/s
26.9
tokens/s
387
ms
pp4096+tg256
2367
tokens/s
19.6
tokens/s
1.78
sec
pp2048+tg256
2718
tokens/s
23.9
tokens/s
794
ms
pp2048+tg768
2713
tokens/s
22.0
tokens/s
801
ms
pp1024+tg1024
2926
tokens/s
23.5
tokens/s
389
ms
pp1280+tg3072
2765
tokens/s
20.6
tokens/s
504
ms
pp384+tg1152
2957
tokens/s
25.0
tokens/s
167
ms
pp64+tg1024
985
tokens/s
26.0
tokens/s
100
ms
pp16+tg1536
267
tokens/s
25.7
tokens/s
100
ms