TEST #229 RESULTS
04/04/2025 - 2:19 PM
ACCELERATOR
MODEL
67.3
tokens/s
generation
651
ms
time to first token
1974
tokens/s
prompt
589
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 5600 6-Core Processor (znver3)
RAM
63.9GB
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
2348
tokens/s
70.3
tokens/s
451
ms
pp4096+tg256
1909
tokens/s
63.6
tokens/s
2.16
sec
pp2048+tg256
2155
tokens/s
66.7
tokens/s
965
ms
pp2048+tg768
2156
tokens/s
67.2
tokens/s
965
ms
pp1024+tg1024
2309
tokens/s
68.1
tokens/s
461
ms
pp1280+tg3072
2239
tokens/s
65.6
tokens/s
587
ms
pp384+tg1152
2385
tokens/s
68.3
tokens/s
175
ms
pp64+tg1024
1702
tokens/s
68.1
tokens/s
53
ms
pp16+tg1536
562
tokens/s
68.1
tokens/s
43
ms