TEST #3573 RESULTS
05/19/2026 - 10:28 AM
ACCELERATOR
MODEL
41.9
tokens/s
generation
1.02
sec
time to first token
1176
tokens/s
prompt
364
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
63.8GB
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
1775
tokens/s
50.4
tokens/s
597
ms
pp4096+tg256
1481
tokens/s
45.3
tokens/s
2.79
sec
pp2048+tg256
1238
tokens/s
46.2
tokens/s
1.68
sec
pp2048+tg768
1225
tokens/s
42.1
tokens/s
1.69
sec
pp1024+tg1024
1230
tokens/s
41.6
tokens/s
856
ms
pp1280+tg3072
1221
tokens/s
34.5
tokens/s
1.07
sec
pp384+tg1152
1210
tokens/s
36.6
tokens/s
343
ms
pp64+tg1024
883
tokens/s
39.9
tokens/s
98
ms
pp16+tg1536
318
tokens/s
40.4
tokens/s
76
ms