TEST #3871 RESULTS
07/07/2026 - 2:03 PM
ACCELERATOR
MODEL
23.4
tokens/s
generation
14.24
sec
time to first token
98
tokens/s
prompt
54
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
13th Gen Intel Core i7-13800H (alderlake)
RAM
31.6GB
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
103
tokens/s
32.1
tokens/s
9.96
sec
pp4096+tg256
85
tokens/s
24.1
tokens/s
48.04
sec
pp2048+tg256
98
tokens/s
28.5
tokens/s
21.00
sec
pp2048+tg768
97
tokens/s
20.5
tokens/s
21.10
sec
pp1024+tg1024
99
tokens/s
25.5
tokens/s
10.41
sec
pp1280+tg3072
100
tokens/s
23.5
tokens/s
12.87
sec
pp384+tg1152
99
tokens/s
26.4
tokens/s
3.90
sec
pp64+tg1024
105
tokens/s
27.5
tokens/s
644
ms
pp16+tg1536
94
tokens/s
2.7
tokens/s
212
ms