TEST #96 RESULTS
03/28/2025 - 9:29 PM
ACCELERATOR
MODEL
28.1
tokens/s
generation
471
ms
time to first token
2936
tokens/s
prompt
560
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 9354 32-Core Processor (znver4)
RAM
755.5GB
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
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
4000
tokens/s
29.7
tokens/s
293
ms
pp4096+tg256
2536
tokens/s
24.2
tokens/s
1.66
sec
pp2048+tg256
3315
tokens/s
27.3
tokens/s
656
ms
pp2048+tg768
3298
tokens/s
26.4
tokens/s
659
ms
pp1024+tg1024
4022
tokens/s
28.1
tokens/s
292
ms
pp1280+tg3072
3629
tokens/s
25.9
tokens/s
389
ms
pp384+tg1152
3988
tokens/s
29.8
tokens/s
130
ms
pp64+tg1024
1264
tokens/s
31.0
tokens/s
82
ms
pp16+tg1536
368
tokens/s
30.4
tokens/s
74
ms