TEST #1819 RESULTS
11/17/2025 - 6:46 AM
ACCELERATOR
MODEL
157
tokens/s
generation
222
ms
time to first token
6254
tokens/s
prompt
1642
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 3600 6-Core Processor (znver2)
RAM
47.9GB
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
7800
tokens/s
152
tokens/s
138
ms
pp4096+tg256
5046
tokens/s
148
tokens/s
819
ms
pp2048+tg256
6624
tokens/s
158
tokens/s
316
ms
pp2048+tg768
6619
tokens/s
160
tokens/s
316
ms
pp1024+tg1024
7864
tokens/s
160
tokens/s
137
ms
pp1280+tg3072
7390
tokens/s
158
tokens/s
180
ms
pp384+tg1152
8870
tokens/s
159
tokens/s
50
ms
pp64+tg1024
4950
tokens/s
159
tokens/s
19
ms
pp16+tg1536
1119
tokens/s
156
tokens/s
21
ms