TEST #327 RESULTS
04/06/2025 - 12:36 AM
ACCELERATOR
MODEL
49.1
tokens/s
generation
603
ms
time to first token
2680
tokens/s
prompt
602
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 3B Instruct - Q8_0
SYSTEM
CPU
AMD Ryzen 9 5950X 16-Core Processor (znver3)
RAM
125.7GB
OS
Linux
Kernel Release
6.13.9-200.fc41.x86_64
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Sat Mar 29 01:29:31 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
3463
tokens/s
61.2
tokens/s
312
ms
pp4096+tg256
1656
tokens/s
10.1
tokens/s
2.57
sec
pp2048+tg256
2597
tokens/s
46.1
tokens/s
810
ms
pp2048+tg768
2591
tokens/s
42.9
tokens/s
812
ms
pp1024+tg1024
3330
tokens/s
53.1
tokens/s
324
ms
pp1280+tg3072
3155
tokens/s
31.3
tokens/s
424
ms
pp384+tg1152
4116
tokens/s
61.6
tokens/s
107
ms
pp64+tg1024
2335
tokens/s
70.2
tokens/s
40
ms
pp16+tg1536
879
tokens/s
65.7
tokens/s
30
ms