TEST #4132 RESULTS
09/08/2026 - 5:11 AM
ACCELERATOR
MODEL
62.9
tokens/s
generation
2.46
sec
time to first token
555
tokens/s
prompt
242
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 9800X3D 8-Core Processor
RAM
30.9GB
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
586
tokens/s
64.6
tokens/s
1.76
sec
pp4096+tg256
481
tokens/s
52.7
tokens/s
8.54
sec
pp2048+tg256
635
tokens/s
63.2
tokens/s
3.24
sec
pp2048+tg768
591
tokens/s
60.6
tokens/s
3.48
sec
pp1024+tg1024
544
tokens/s
64.0
tokens/s
1.90
sec
pp1280+tg3072
540
tokens/s
58.8
tokens/s
2.39
sec
pp384+tg1152
572
tokens/s
66.9
tokens/s
686
ms
pp64+tg1024
565
tokens/s
68.3
tokens/s
128
ms
pp16+tg1536
479
tokens/s
67.3
tokens/s
48
ms