TEST #1771 RESULTS
11/09/2025 - 10:11 PM
ACCELERATOR
MODEL
112
tokens/s
generation
205
ms
time to first token
7529
tokens/s
prompt
1604
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
RAM
119.7GB
OS
Linux
Kernel Release
6.11.0-1016-nvidia
Architecture
aarch64
Version
Cosmopolitan 3.9.7 MODE=aarch64; #16-Ubuntu SMP PREEMPT_DYNAMIC Sun Sep 21 16:52:46 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
9222
tokens/s
104
tokens/s
119
ms
pp4096+tg256
5172
tokens/s
94.1
tokens/s
804
ms
pp2048+tg256
7361
tokens/s
104
tokens/s
290
ms
pp2048+tg768
7405
tokens/s
105
tokens/s
288
ms
pp1024+tg1024
9556
tokens/s
122
tokens/s
116
ms
pp1280+tg3072
8828
tokens/s
114
tokens/s
153
ms
pp384+tg1152
12787
tokens/s
119
tokens/s
41
ms
pp64+tg1024
5700
tokens/s
126
tokens/s
18
ms
pp16+tg1536
1732
tokens/s
123
tokens/s
16
ms