TEST #3179 RESULTS
03/24/2026 - 11:48 PM
ACCELERATOR
MODEL
3.5
tokens/s
generation
11.86
sec
time to first token
110
tokens/s
prompt
32
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 8845HS w/ Radeon 780M Graphics (znver4)
RAM
23.3GB
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
119
tokens/s
4.1
tokens/s
8.88
sec
pp4096+tg256
102
tokens/s
0.8
tokens/s
41.46
sec
pp2048+tg256
128
tokens/s
2.9
tokens/s
16.34
sec
pp2048+tg768
124
tokens/s
2.5
tokens/s
16.92
sec
pp1024+tg1024
133
tokens/s
3.6
tokens/s
7.94
sec
pp1280+tg3072
127
tokens/s
1.8
tokens/s
10.35
sec
pp384+tg1152
124
tokens/s
4.7
tokens/s
3.27
sec
pp64+tg1024
56
tokens/s
5.9
tokens/s
1.28
sec
pp16+tg1536
76
tokens/s
5.2
tokens/s
333
ms