TEST #3678 RESULTS
06/03/2026 - 2:40 PM
ACCELERATOR
MODEL
158
tokens/s
generation
193
ms
time to first token
7964
tokens/s
prompt
1868
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
RAM
121.6GB
OS
Linux
Kernel Release
6.17.0-1018-nvidia
Architecture
aarch64
Version
Cosmopolitan 3.9.7 MODE=aarch64; #18-Ubuntu SMP PREEMPT_DYNAMIC Tue May 5 21:28:33 UTC 2026
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
9885
tokens/s
165
tokens/s
110
ms
pp4096+tg256
5390
tokens/s
136
tokens/s
767
ms
pp2048+tg256
7771
tokens/s
154
tokens/s
270
ms
pp2048+tg768
7759
tokens/s
151
tokens/s
270
ms
pp1024+tg1024
9967
tokens/s
160
tokens/s
109
ms
pp1280+tg3072
9396
tokens/s
148
tokens/s
142
ms
pp384+tg1152
13318
tokens/s
167
tokens/s
35
ms
pp64+tg1024
6282
tokens/s
171
tokens/s
16
ms
pp16+tg1536
1903
tokens/s
168
tokens/s
14
ms