TEST #4276 RESULTS
10/13/2026 - 10:58 AM
ACCELERATOR
MODEL
162
tokens/s
generation
251
ms
time to first token
5959
tokens/s
prompt
1568
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
12th Gen Intel Core i7-12650H (alderlake)
RAM
15.3GB
OS
Linux
Kernel Release
7.2.5-4-omarchy
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Wed, 16 Sep 2026 04:45:18 +0000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
7129
tokens/s
182
tokens/s
149
ms
pp4096+tg256
4282
tokens/s
136
tokens/s
964
ms
pp2048+tg256
5819
tokens/s
157
tokens/s
358
ms
pp2048+tg768
5768
tokens/s
153
tokens/s
361
ms
pp1024+tg1024
7029
tokens/s
164
tokens/s
151
ms
pp1280+tg3072
6682
tokens/s
148
tokens/s
197
ms
pp384+tg1152
8777
tokens/s
171
tokens/s
49
ms
pp64+tg1024
5902
tokens/s
178
tokens/s
16
ms
pp16+tg1536
2240
tokens/s
173
tokens/s
12
ms