TEST #1588 RESULTS
10/05/2025 - 2:26 PM
ACCELERATOR
MODEL
62.7
tokens/s
generation
2.96
sec
time to first token
507
tokens/s
prompt
221
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 7700 8-Core Processor (znver4)
RAM
31.2GB
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
538
tokens/s
64.5
tokens/s
1.92
sec
pp4096+tg256
388
tokens/s
53.7
tokens/s
10.58
sec
pp2048+tg256
483
tokens/s
61.6
tokens/s
4.25
sec
pp2048+tg768
462
tokens/s
60.2
tokens/s
4.45
sec
pp1024+tg1024
501
tokens/s
64.0
tokens/s
2.06
sec
pp1280+tg3072
521
tokens/s
57.9
tokens/s
2.47
sec
pp384+tg1152
527
tokens/s
66.4
tokens/s
743
ms
pp64+tg1024
551
tokens/s
68.4
tokens/s
130
ms
pp16+tg1536
593
tokens/s
67.2
tokens/s
41
ms