TEST #1571 RESULTS
09/29/2025 - 11:04 PM
ACCELERATOR
MODEL
151
tokens/s
generation
284
ms
time to first token
4403
tokens/s
prompt
1327
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 5900X 12-Core Processor (znver3)
RAM
63.9GB
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
5777
tokens/s
172
tokens/s
184
ms
pp4096+tg256
4152
tokens/s
128
tokens/s
994
ms
pp2048+tg256
5156
tokens/s
118
tokens/s
406
ms
pp2048+tg768
5171
tokens/s
111
tokens/s
405
ms
pp1024+tg1024
5663
tokens/s
144
tokens/s
188
ms
pp1280+tg3072
5345
tokens/s
106
tokens/s
246
ms
pp384+tg1152
5531
tokens/s
176
tokens/s
74
ms
pp64+tg1024
1587
tokens/s
211
tokens/s
45
ms
pp16+tg1536
1247
tokens/s
191
tokens/s
17
ms