TEST #3416 RESULTS
04/29/2026 - 11:51 AM
ACCELERATOR
MODEL
25.8
tokens/s
generation
1.66
sec
time to first token
797
tokens/s
prompt
231
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i7-7700T CPU @ 2.90GHz (skylake)
RAM
47.4GB
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
932
tokens/s
27.3
tokens/s
1.14
sec
pp4096+tg256
734
tokens/s
23.6
tokens/s
5.62
sec
pp2048+tg256
849
tokens/s
25.7
tokens/s
2.45
sec
pp2048+tg768
848
tokens/s
25.2
tokens/s
2.45
sec
pp1024+tg1024
913
tokens/s
26.2
tokens/s
1.16
sec
pp1280+tg3072
882
tokens/s
24.7
tokens/s
1.49
sec
pp384+tg1152
975
tokens/s
26.3
tokens/s
430
ms
pp64+tg1024
786
tokens/s
27.0
tokens/s
117
ms
pp16+tg1536
254
tokens/s
26.7
tokens/s
99
ms