TEST #58 RESULTS
03/28/2025 - 8:48 PM
ACCELERATOR
MODEL
50.9
tokens/s
generation
846
ms
time to first token
1524
tokens/s
prompt
451
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7B13 64-Core Processor (znver3)
RAM
251.5GB
OS
Linux
Kernel Release
6.8.0-48-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #48~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Mon Oct 7 11:24:13 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1762
tokens/s
52.7
tokens/s
600
ms
pp4096+tg256
1486
tokens/s
45.2
tokens/s
2.78
sec
pp2048+tg256
1664
tokens/s
49.9
tokens/s
1.25
sec
pp2048+tg768
1665
tokens/s
49.2
tokens/s
1.25
sec
pp1024+tg1024
1738
tokens/s
51.7
tokens/s
608
ms
pp1280+tg3072
1727
tokens/s
48.2
tokens/s
760
ms
pp384+tg1152
1789
tokens/s
53.0
tokens/s
251
ms
pp64+tg1024
1428
tokens/s
54.2
tokens/s
63
ms
pp16+tg1536
458
tokens/s
53.6
tokens/s
52
ms