TEST #3365 RESULTS
04/22/2026 - 9:48 PM
ACCELERATOR
MODEL
44.9
tokens/s
generation
619
ms
time to first token
2182
tokens/s
prompt
541
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i5-14400F (alderlake)
RAM
15.8GB
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
2677
tokens/s
45.9
tokens/s
404
ms
pp4096+tg256
1890
tokens/s
43.0
tokens/s
2.19
sec
pp2048+tg256
2345
tokens/s
45.1
tokens/s
896
ms
pp2048+tg768
2341
tokens/s
44.9
tokens/s
897
ms
pp1024+tg1024
2674
tokens/s
45.1
tokens/s
405
ms
pp1280+tg3072
2606
tokens/s
44.0
tokens/s
514
ms
pp384+tg1152
3046
tokens/s
45.3
tokens/s
151
ms
pp64+tg1024
1559
tokens/s
45.3
tokens/s
63
ms
pp16+tg1536
500
tokens/s
45.2
tokens/s
54
ms