TEST #2011 RESULTS
12/21/2025 - 1:25 PM
ACCELERATOR
MODEL
186
tokens/s
generation
189
ms
time to first token
7096
tokens/s
prompt
1911
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
31.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
8750
tokens/s
167
tokens/s
124
ms
pp4096+tg256
6050
tokens/s
157
tokens/s
684
ms
pp2048+tg256
7855
tokens/s
189
tokens/s
267
ms
pp2048+tg768
7862
tokens/s
183
tokens/s
266
ms
pp1024+tg1024
9086
tokens/s
196
tokens/s
119
ms
pp1280+tg3072
8615
tokens/s
152
tokens/s
158
ms
pp384+tg1152
9812
tokens/s
205
tokens/s
46
ms
pp64+tg1024
4951
tokens/s
214
tokens/s
19
ms
pp16+tg1536
882
tokens/s
214
tokens/s
23
ms