TEST #328 RESULTS
04/06/2025 - 12:36 AM
ACCELERATOR
MODEL
52.6
tokens/s
generation
204
ms
time to first token
6655
tokens/s
prompt
1197
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 7800X3D 8-Core Processor (znver4)
RAM
31.6GB
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
9220
tokens/s
53.0
tokens/s
129
ms
pp4096+tg256
6505
tokens/s
48.9
tokens/s
650
ms
pp2048+tg256
8175
tokens/s
53.7
tokens/s
270
ms
pp2048+tg768
8268
tokens/s
52.7
tokens/s
266
ms
pp1024+tg1024
9220
tokens/s
52.7
tokens/s
149
ms
pp1280+tg3072
8174
tokens/s
52.2
tokens/s
175
ms
pp384+tg1152
8817
tokens/s
52.7
tokens/s
65
ms
pp64+tg1024
993
tokens/s
53.8
tokens/s
86
ms
pp16+tg1536
524
tokens/s
53.9
tokens/s
48
ms