TEST #788 RESULTS
04/30/2025 - 3:53 PM
ACCELERATOR
MODEL
29.1
tokens/s
generation
13.10
sec
time to first token
106
tokens/s
prompt
62
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
11th Gen Intel Core i5-11400 @ 2.60GHz (rocketlake)
RAM
31.6GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
114
tokens/s
32.6
tokens/s
9.00
sec
pp4096+tg256
91
tokens/s
23.9
tokens/s
45.27
sec
pp2048+tg256
105
tokens/s
29.4
tokens/s
19.46
sec
pp2048+tg768
106
tokens/s
28.5
tokens/s
19.35
sec
pp1024+tg1024
113
tokens/s
31.1
tokens/s
9.11
sec
pp1280+tg3072
112
tokens/s
26.6
tokens/s
11.47
sec
pp384+tg1152
119
tokens/s
29.3
tokens/s
3.25
sec
pp64+tg1024
89
tokens/s
28.9
tokens/s
764
ms
pp16+tg1536
105
tokens/s
31.4
tokens/s
183
ms