TEST #37 RESULTS
03/28/2025 - 6:08 PM
ACCELERATOR
MODEL
112
tokens/s
generation
112
ms
time to first token
12246
tokens/s
prompt
2305
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7352 24-Core Processor (znver2)
RAM
127.9GB
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
15769
tokens/s
110
tokens/s
74
ms
pp4096+tg256
11554
tokens/s
117
tokens/s
363
ms
pp2048+tg256
14311
tokens/s
110
tokens/s
161
ms
pp2048+tg768
14452
tokens/s
96.5
tokens/s
159
ms
pp1024+tg1024
16419
tokens/s
114
tokens/s
71
ms
pp1280+tg3072
15400
tokens/s
115
tokens/s
92
ms
pp384+tg1152
16764
tokens/s
114
tokens/s
32
ms
pp64+tg1024
4692
tokens/s
113
tokens/s
22
ms
pp16+tg1536
854
tokens/s
114
tokens/s
28
ms