TEST #1268 RESULTS
07/29/2025 - 10:03 PM
ACCELERATOR
MODEL
72.5
tokens/s
generation
325
ms
time to first token
3868
tokens/s
prompt
952
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
12th Gen Intel Core i7-12800H (alderlake)
RAM
63.7GB
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
7447
tokens/s
122
tokens/s
146
ms
pp4096+tg256
4785
tokens/s
123
tokens/s
866
ms
pp2048+tg256
5950
tokens/s
132
tokens/s
352
ms
pp2048+tg768
5609
tokens/s
112
tokens/s
375
ms
pp1024+tg1024
5026
tokens/s
70.6
tokens/s
216
ms
pp1280+tg3072
4051
tokens/s
22.2
tokens/s
326
ms
pp384+tg1152
1000
tokens/s
22.0
tokens/s
422
ms
pp64+tg1024
685
tokens/s
24.5
tokens/s
129
ms
pp16+tg1536
260
tokens/s
24.5
tokens/s
96
ms