From e7f2a8beb75dedbd1a3ef89b3a4315c9dcdd7b9c Mon Sep 17 00:00:00 2001 From: Maximo Piva Date: Tue, 27 Aug 2024 01:39:37 -0300 Subject: [PATCH] Update README.md --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 703245a..4ed4f08 100644 --- a/README.md +++ b/README.md @@ -78,7 +78,7 @@ _Since the program can take any chainner save file as a parameter you're open to The current chainner file was an experimentation by me, looking for a good compromise in artifact removal without texture crushing. But new models are created every day, and other can come with better chainner files. -I found onnx models seems to be faster that pytorch and NCNN models. (In my machine). +I found onnx models seems to be faster than pytorch and NCNN models. (RTX 4070). Most of the time you can convert the pytorch model to ONNX or to NCNN using chainner. Follow chaiNNer guides.