TEST #3265 RESULTS
04/09/2026 - 6:39 PM
ACCELERATOR
MODEL
155
tokens/s
generation
301
ms
time to first token
4809
tokens/s
prompt
1353
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 3 3100 4-Core Processor (znver2)
RAM
31.2GB
OS
Linux
Kernel Release
6.18.7-76061807-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #202601231045~1769703228~24.04~cb87b5b SMP PREEMPT_DYNAMIC Thu J
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
5636
tokens/s
169
tokens/s
188
ms
pp4096+tg256
3706
tokens/s
128
tokens/s
1.11
sec
pp2048+tg256
4784
tokens/s
151
tokens/s
435
ms
pp2048+tg768
4784
tokens/s
146
tokens/s
435
ms
pp1024+tg1024
5632
tokens/s
159
tokens/s
188
ms
pp1280+tg3072
5255
tokens/s
140
tokens/s
250
ms
pp384+tg1152
6760
tokens/s
165
tokens/s
63
ms
pp64+tg1024
5691
tokens/s
171
tokens/s
16
ms
pp16+tg1536
1036
tokens/s
166
tokens/s
20
ms