TEST #491 RESULTS
04/07/2025 - 11:52 AM
ACCELERATOR
MODEL
152
tokens/s
generation
278
ms
time to first token
4587
tokens/s
prompt
1357
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 5800X3D 8-Core Processor (znver3)
RAM
31.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
6124
tokens/s
177
tokens/s
173
ms
pp4096+tg256
4123
tokens/s
128
tokens/s
1.00
sec
pp2048+tg256
5282
tokens/s
121
tokens/s
396
ms
pp2048+tg768
5282
tokens/s
112
tokens/s
396
ms
pp1024+tg1024
6012
tokens/s
145
tokens/s
176
ms
pp1280+tg3072
5652
tokens/s
107
tokens/s
233
ms
pp384+tg1152
5839
tokens/s
177
tokens/s
70
ms
pp64+tg1024
1651
tokens/s
208
tokens/s
42
ms
pp16+tg1536
1314
tokens/s
188
tokens/s
16
ms