TEST #4237 RESULTS
10/01/2026 - 11:49 AM
ACCELERATOR
MODEL
37.2
tokens/s
generation
10.49
sec
time to first token
133
tokens/s
prompt
78
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
12th Gen Intel Core i5-12500 (alderlake)
RAM
15.7GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
152
tokens/s
42.8
tokens/s
6.74
sec
pp4096+tg256
119
tokens/s
31.8
tokens/s
34.39
sec
pp2048+tg256
128
tokens/s
38.4
tokens/s
16.00
sec
pp2048+tg768
128
tokens/s
37.2
tokens/s
16.05
sec
pp1024+tg1024
130
tokens/s
36.7
tokens/s
7.90
sec
pp1280+tg3072
132
tokens/s
32.3
tokens/s
9.72
sec
pp384+tg1152
129
tokens/s
37.4
tokens/s
3.01
sec
pp64+tg1024
140
tokens/s
40.0
tokens/s
479
ms
pp16+tg1536
140
tokens/s
37.9
tokens/s
136
ms