TEST #357 RESULTS
04/06/2025 - 6:08 AM
ACCELERATOR
MODEL
161
tokens/s
generation
126
ms
time to first token
10794
tokens/s
prompt
2395
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen Threadripper PRO 3955WX 16-Cores (znver2)
RAM
127.8GB
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
13640
tokens/s
157
tokens/s
81
ms
pp4096+tg256
9235
tokens/s
159
tokens/s
450
ms
pp2048+tg256
12058
tokens/s
163
tokens/s
176
ms
pp2048+tg768
11893
tokens/s
157
tokens/s
179
ms
pp1024+tg1024
14015
tokens/s
159
tokens/s
79
ms
pp1280+tg3072
13228
tokens/s
164
tokens/s
104
ms
pp384+tg1152
15075
tokens/s
164
tokens/s
33
ms
pp64+tg1024
6806
tokens/s
162
tokens/s
15
ms
pp16+tg1536
1200
tokens/s
162
tokens/s
20
ms