TEST #3300 RESULTS
04/13/2026 - 8:55 AM
ACCELERATOR
MODEL
35.3
tokens/s
generation
8.97
sec
time to first token
159
tokens/s
prompt
85
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
12th Gen Intel Core i7-12700 (alderlake)
RAM
63.8GB
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
165
tokens/s
36.3
tokens/s
6.22
sec
pp4096+tg256
132
tokens/s
31.8
tokens/s
30.97
sec
pp2048+tg256
154
tokens/s
34.8
tokens/s
13.33
sec
pp2048+tg768
153
tokens/s
34.3
tokens/s
13.39
sec
pp1024+tg1024
165
tokens/s
36.0
tokens/s
6.23
sec
pp1280+tg3072
163
tokens/s
33.7
tokens/s
7.87
sec
pp384+tg1152
179
tokens/s
35.7
tokens/s
2.17
sec
pp64+tg1024
188
tokens/s
37.6
tokens/s
367
ms
pp16+tg1536
128
tokens/s
37.4
tokens/s
150
ms