TEST #3894 RESULTS
07/12/2026 - 1:21 PM
ACCELERATOR
MODEL
112
tokens/s
generation
1.17
sec
time to first token
1238
tokens/s
prompt
491
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD RYZEN AI MAX+ 395 w/ Radeon 8060S
RAM
62.4GB
OS
Linux
Kernel Release
6.14.0-24-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #24~24.04.3-Ubuntu SMP PREEMPT_DYNAMIC Mon Jul 7 16:39:17 UTC 2
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1336
tokens/s
118
tokens/s
776
ms
pp4096+tg256
963
tokens/s
94.7
tokens/s
4.26
sec
pp2048+tg256
1218
tokens/s
109
tokens/s
1.69
sec
pp2048+tg768
1208
tokens/s
107
tokens/s
1.71
sec
pp1024+tg1024
1353
tokens/s
114
tokens/s
766
ms
pp1280+tg3072
1330
tokens/s
104
tokens/s
973
ms
pp384+tg1152
1454
tokens/s
119
tokens/s
274
ms
pp64+tg1024
1461
tokens/s
122
tokens/s
53
ms
pp16+tg1536
822
tokens/s
120
tokens/s
28
ms