TEST #3467 RESULTS
05/07/2026 - 10:40 PM
ACCELERATOR
MODEL
117
tokens/s
generation
1.09
sec
time to first token
2778
tokens/s
prompt
667
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 5950X 16-Core Processor (znver3)
RAM
127.9GB
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
3770
tokens/s
145
tokens/s
279
ms
pp4096+tg256
1566
tokens/s
53.9
tokens/s
4.47
sec
pp2048+tg256
3210
tokens/s
102
tokens/s
648
ms
pp2048+tg768
3229
tokens/s
96.0
tokens/s
644
ms
pp1024+tg1024
3772
tokens/s
124
tokens/s
279
ms
pp1280+tg3072
3592
tokens/s
90.3
tokens/s
364
ms
pp384+tg1152
3857
tokens/s
150
tokens/s
105
ms
pp64+tg1024
61
tokens/s
129
tokens/s
3.03
sec
pp16+tg1536
1943
tokens/s
160
tokens/s
13
ms