TEST #3987 RESULTS
08/03/2026 - 11:40 AM
ACCELERATOR
MODEL
19.9
tokens/s
generation
19.30
sec
time to first token
71
tokens/s
prompt
42
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i7-6820HQ CPU @ 2.70GHz (skylake)
RAM
31.9GB
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
83
tokens/s
23.8
tokens/s
12.37
sec
pp4096+tg256
65
tokens/s
17.3
tokens/s
63.17
sec
pp2048+tg256
72
tokens/s
20.7
tokens/s
28.52
sec
pp2048+tg768
64
tokens/s
18.0
tokens/s
32.23
sec
pp1024+tg1024
70
tokens/s
20.4
tokens/s
14.74
sec
pp1280+tg3072
76
tokens/s
20.0
tokens/s
16.97
sec
pp384+tg1152
86
tokens/s
24.8
tokens/s
4.50
sec
pp64+tg1024
81
tokens/s
17.3
tokens/s
828
ms
pp16+tg1536
45
tokens/s
16.9
tokens/s
415
ms