TEST #2131 RESULTS
01/04/2026 - 6:35 AM
ACCELERATOR
MODEL
44.5
tokens/s
generation
944
ms
time to first token
1426
tokens/s
prompt
407
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
11th Gen Intel Core i7-11800H @ 2.30GHz (tigerlake)
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
1720
tokens/s
47.0
tokens/s
618
ms
pp4096+tg256
1212
tokens/s
36.7
tokens/s
3.42
sec
pp2048+tg256
1574
tokens/s
46.3
tokens/s
1.32
sec
pp2048+tg768
1575
tokens/s
45.6
tokens/s
1.32
sec
pp1024+tg1024
1698
tokens/s
47.2
tokens/s
625
ms
pp1280+tg3072
1604
tokens/s
42.3
tokens/s
820
ms
pp384+tg1152
1815
tokens/s
45.5
tokens/s
232
ms
pp64+tg1024
1306
tokens/s
45.9
tokens/s
70
ms
pp16+tg1536
331
tokens/s
44.4
tokens/s
69
ms