TEST #1681 RESULTS
10/18/2025 - 7:57 PM
ACCELERATOR
MODEL
6.1
tokens/s
generation
1.52
sec
time to first token
1017
tokens/s
prompt
160
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 7700 8-Core Processor (znver4)
RAM
63.1GB
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
1294
tokens/s
6.6
tokens/s
943
ms
pp4096+tg256
791
tokens/s
2.0
tokens/s
5.68
sec
pp2048+tg256
1084
tokens/s
4.4
tokens/s
2.15
sec
pp2048+tg768
1087
tokens/s
4.0
tokens/s
2.14
sec
pp1024+tg1024
1290
tokens/s
5.8
tokens/s
946
ms
pp1280+tg3072
1185
tokens/s
3.4
tokens/s
1.26
sec
pp384+tg1152
1407
tokens/s
8.2
tokens/s
358
ms
pp64+tg1024
770
tokens/s
11.4
tokens/s
133
ms
pp16+tg1536
239
tokens/s
9.4
tokens/s
109
ms