TEST #1438 RESULTS
08/28/2025 - 1:29 PM
ACCELERATOR
MODEL
183
tokens/s
generation
146
ms
time to first token
10193
tokens/s
prompt
2338
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core Ultra 7 265 (arrowlake-s)
RAM
31.5GB
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
12839
tokens/s
186
tokens/s
85
ms
pp4096+tg256
7279
tokens/s
176
tokens/s
569
ms
pp2048+tg256
10347
tokens/s
153
tokens/s
205
ms
pp2048+tg768
10454
tokens/s
184
tokens/s
202
ms
pp1024+tg1024
13410
tokens/s
188
tokens/s
83
ms
pp1280+tg3072
12156
tokens/s
184
tokens/s
112
ms
pp384+tg1152
16846
tokens/s
190
tokens/s
29
ms
pp64+tg1024
6737
tokens/s
195
tokens/s
16
ms
pp16+tg1536
1672
tokens/s
194
tokens/s
17
ms