TEST #1487 RESULTS
09/05/2025 - 1:11 PM
ACCELERATOR
MODEL
131
tokens/s
generation
237
ms
time to first token
5884
tokens/s
prompt
1481
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core Ultra 7 265H (arrowlake)
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
7629
tokens/s
137
tokens/s
142
ms
pp4096+tg256
4660
tokens/s
124
tokens/s
890
ms
pp2048+tg256
6450
tokens/s
134
tokens/s
325
ms
pp2048+tg768
6177
tokens/s
125
tokens/s
341
ms
pp1024+tg1024
7533
tokens/s
132
tokens/s
145
ms
pp1280+tg3072
6889
tokens/s
126
tokens/s
193
ms
pp384+tg1152
9594
tokens/s
133
tokens/s
47
ms
pp64+tg1024
2539
tokens/s
134
tokens/s
33
ms
pp16+tg1536
1487
tokens/s
133
tokens/s
18
ms