TEST #2 RESULTS
03/19/2025 - 9:53 PM
ACCELERATOR
MODEL
59.9
tokens/s
generation
640
ms
time to first token
2005
tokens/s
prompt
573
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7352 24-Core Processor (znver2)
RAM
125.64GB
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
0.9.1
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2370
tokens/s
62.3
tokens/s
449
ms
pp4096+tg256
1963
tokens/s
53.7
tokens/s
2.11
sec
pp2048+tg256
2216
tokens/s
59.1
tokens/s
942
ms
pp2048+tg768
2215
tokens/s
58.1
tokens/s
942
ms
pp1024+tg1024
2358
tokens/s
60.6
tokens/s
451
ms
pp1280+tg3072
2283
tokens/s
57.1
tokens/s
577
ms
pp384+tg1152
2381
tokens/s
62.3
tokens/s
196
ms
pp64+tg1024
1796
tokens/s
63.5
tokens/s
51
ms
pp16+tg1536
467
tokens/s
62.9
tokens/s
49
ms