TEST #1347 RESULTS
08/09/2025 - 2:12 PM
ACCELERATOR
MODEL
10.7
tokens/s
generation
2.29
sec
time to first token
565
tokens/s
prompt
138
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 PRO 8840HS w/ Radeon 780M Graphics (znver4)
RAM
61.8GB
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
703
tokens/s
12.1
tokens/s
1.53
sec
pp4096+tg256
532
tokens/s
3.4
tokens/s
7.99
sec
pp2048+tg256
648
tokens/s
7.9
tokens/s
3.32
sec
pp2048+tg768
628
tokens/s
7.2
tokens/s
3.40
sec
pp1024+tg1024
695
tokens/s
10.3
tokens/s
1.55
sec
pp1280+tg3072
725
tokens/s
6.0
tokens/s
1.86
sec
pp384+tg1152
667
tokens/s
14.3
tokens/s
624
ms
pp64+tg1024
265
tokens/s
19.3
tokens/s
271
ms
pp16+tg1536
224
tokens/s
16.1
tokens/s
103
ms