TEST #630 RESULTS
04/09/2025 - 7:28 AM
ACCELERATOR
MODEL
84.2
tokens/s
generation
189
ms
time to first token
7015
tokens/s
prompt
1461
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 3900X 12-Core Processor (znver2)
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
9154
tokens/s
77.4
tokens/s
125
ms
pp4096+tg256
6329
tokens/s
76.2
tokens/s
661
ms
pp2048+tg256
8180
tokens/s
86.7
tokens/s
263
ms
pp2048+tg768
8165
tokens/s
87.8
tokens/s
263
ms
pp1024+tg1024
9301
tokens/s
88.2
tokens/s
122
ms
pp1280+tg3072
9029
tokens/s
86.0
tokens/s
153
ms
pp384+tg1152
8347
tokens/s
86.1
tokens/s
58
ms
pp64+tg1024
3569
tokens/s
84.6
tokens/s
30
ms
pp16+tg1536
1066
tokens/s
84.8
tokens/s
27
ms