TEST #1196 RESULTS
07/08/2025 - 7:07 PM
ACCELERATOR
MODEL
151
tokens/s
generation
118
ms
time to first token
11382
tokens/s
prompt
2445
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen Threadripper 3970X 32-Core Processor (znver2)
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
14447
tokens/s
161
tokens/s
78
ms
pp4096+tg256
10281
tokens/s
160
tokens/s
405
ms
pp2048+tg256
12357
tokens/s
161
tokens/s
172
ms
pp2048+tg768
13038
tokens/s
155
tokens/s
163
ms
pp1024+tg1024
15128
tokens/s
145
tokens/s
74
ms
pp1280+tg3072
14399
tokens/s
144
tokens/s
95
ms
pp384+tg1152
15759
tokens/s
143
tokens/s
31
ms
pp64+tg1024
5985
tokens/s
145
tokens/s
18
ms
pp16+tg1536
1045
tokens/s
145
tokens/s
23
ms