TEST #1578 RESULTS
10/02/2025 - 10:11 PM
ACCELERATOR
MODEL
3.8
tokens/s
generation
11.29
sec
time to first token
112
tokens/s
prompt
34
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 PRO 7840U w/ Radeon 780M Graphics (znver4)
RAM
27.7GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
133
tokens/s
4.5
tokens/s
7.94
sec
pp4096+tg256
119
tokens/s
0.7
tokens/s
35.95
sec
pp2048+tg256
126
tokens/s
3.2
tokens/s
16.56
sec
pp2048+tg768
122
tokens/s
2.8
tokens/s
17.20
sec
pp1024+tg1024
125
tokens/s
3.9
tokens/s
8.41
sec
pp1280+tg3072
122
tokens/s
1.9
tokens/s
10.73
sec
pp384+tg1152
124
tokens/s
5.1
tokens/s
3.25
sec
pp64+tg1024
54
tokens/s
6.5
tokens/s
1.29
sec
pp16+tg1536
81
tokens/s
5.7
tokens/s
306
ms