TEST #1073 RESULTS
06/18/2025 - 4:06 PM
ACCELERATOR
MODEL
180
tokens/s
generation
156
ms
time to first token
8826
tokens/s
prompt
2167
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 5800X 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
11100
tokens/s
148
tokens/s
101
ms
pp4096+tg256
7278
tokens/s
144
tokens/s
568
ms
pp2048+tg256
9536
tokens/s
148
tokens/s
223
ms
pp2048+tg768
9560
tokens/s
176
tokens/s
221
ms
pp1024+tg1024
11499
tokens/s
201
tokens/s
94
ms
pp1280+tg3072
10750
tokens/s
197
tokens/s
125
ms
pp384+tg1152
12653
tokens/s
204
tokens/s
35
ms
pp64+tg1024
6182
tokens/s
202
tokens/s
15
ms
pp16+tg1536
872
tokens/s
203
tokens/s
24
ms