TEST #727 RESULTS
04/15/2025 - 4:50 PM
ACCELERATOR
MODEL
9.0
tokens/s
generation
11.11
sec
time to first token
129
tokens/s
prompt
47
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 5700X3D 8-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
137
tokens/s
10.8
tokens/s
7.56
sec
pp4096+tg256
105
tokens/s
3.8
tokens/s
39.22
sec
pp2048+tg256
129
tokens/s
6.6
tokens/s
16.08
sec
pp2048+tg768
129
tokens/s
6.0
tokens/s
16.08
sec
pp1024+tg1024
136
tokens/s
8.5
tokens/s
7.60
sec
pp1280+tg3072
129
tokens/s
5.2
tokens/s
10.04
sec
pp384+tg1152
143
tokens/s
11.6
tokens/s
2.75
sec
pp64+tg1024
143
tokens/s
15.3
tokens/s
484
ms
pp16+tg1536
110
tokens/s
13.0
tokens/s
180
ms