RTNeural icon indicating copy to clipboard operation
RTNeural copied to clipboard

Published 20 hours ago • verified publisher icon jatinchowdhury18

Comparison of this framework’s performance with other inference frameworks:

Open yudashuixiao1 opened this issue 2 months ago • 1 comments

Has anyone compared this framework’s runtime efficiency with TensorFlow Lite, MNN, or NCNN?

yudashuixiao1 avatar Sep 09 '25 06:09 yudashuixiao1

The RTNeural-compare repository has some performance comparisons between RTNeural, libtorch, TensorFlow Lite, and the ONNX runtime, mostly broken down by layer type and size. Comparisons with additional inference engines would definitely be nice, if you or anyone else is interested in contributing!

jatinchowdhury18 avatar Sep 10 '25 14:09 jatinchowdhury18