TEST #1876 RESULTS
11/27/2025 - 5:22 PM
ACCELERATOR
MODEL
26.8
tokens/s
generation
12.83
sec
time to first token
116
tokens/s
prompt
62
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i5-8500 CPU @ 3.00GHz (skylake)
RAM
31.8GB
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
117
tokens/s
28.7
tokens/s
8.79
sec
pp4096+tg256
90
tokens/s
21.0
tokens/s
45.66
sec
pp2048+tg256
109
tokens/s
25.4
tokens/s
18.86
sec
pp2048+tg768
108
tokens/s
25.3
tokens/s
18.93
sec
pp1024+tg1024
119
tokens/s
27.4
tokens/s
8.61
sec
pp1280+tg3072
118
tokens/s
24.2
tokens/s
10.92
sec
pp384+tg1152
127
tokens/s
29.0
tokens/s
3.06
sec
pp64+tg1024
133
tokens/s
30.1
tokens/s
512
ms
pp16+tg1536
120
tokens/s
29.6
tokens/s
164
ms