TEST #469 RESULTS
04/07/2025 - 4:09 AM
ACCELERATOR
MODEL
60.5
tokens/s
generation
611
ms
time to first token
2160
tokens/s
prompt
598
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 5600X 6-Core Processor (znver3)
RAM
31.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2531
tokens/s
62.2
tokens/s
421
ms
pp4096+tg256
1984
tokens/s
56.1
tokens/s
2.08
sec
pp2048+tg256
2327
tokens/s
59.8
tokens/s
897
ms
pp2048+tg768
2324
tokens/s
59.2
tokens/s
898
ms
pp1024+tg1024
2536
tokens/s
61.0
tokens/s
420
ms
pp1280+tg3072
2463
tokens/s
58.5
tokens/s
536
ms
pp384+tg1152
2628
tokens/s
62.2
tokens/s
162
ms
pp64+tg1024
2010
tokens/s
62.9
tokens/s
47
ms
pp16+tg1536
631
tokens/s
62.5
tokens/s
41
ms