TEST #2275 RESULTS
01/13/2026 - 3:58 AM
ACCELERATOR
MODEL
29.8
tokens/s
generation
9.51
sec
time to first token
159
tokens/s
prompt
79
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i7-9850H CPU @ 2.60GHz (skylake)
RAM
31.7GB
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
172
tokens/s
31.1
tokens/s
5.97
sec
pp4096+tg256
114
tokens/s
25.2
tokens/s
35.93
sec
pp2048+tg256
161
tokens/s
29.0
tokens/s
12.74
sec
pp2048+tg768
148
tokens/s
28.3
tokens/s
13.88
sec
pp1024+tg1024
167
tokens/s
30.3
tokens/s
6.15
sec
pp1280+tg3072
164
tokens/s
27.3
tokens/s
7.86
sec
pp384+tg1152
151
tokens/s
31.6
tokens/s
2.58
sec
pp64+tg1024
209
tokens/s
32.6
tokens/s
337
ms
pp16+tg1536
147
tokens/s
32.9
tokens/s
136
ms