Stuck on GPU integration with Tensor flow

I have been stuck on getting tnesorflow to work with my GPU.

Here is my environment:
Windows 10 Pro
Tnesorflow 2.6.2
Visual Studio 2022
Nvidia Studio drive 512.15
CUDA 11.6
CuDNN v8.3
cuDNN64_8.dll installed, path C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin
zlibwapi installed in Cuda/11.6/bin/zlibwapi.dll

I have freshly reinstalled all the drives and am confused on setting the paths.

I using the Environment Variable tool under System properties.

SET PATH=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.0\bin;%PATH%
SET PATH=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.0\extras\CUPTI\lib64;%PATH%
SET PATH=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.0\include;%PATH%
SET PATH=C:\tools\cuda\bin;%PATH%

What would the path setting be for each of the above
CUDA 11.6 is at C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6
CUPTI is at C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\extras\CUPTI
Cudnn64_8.dd is at C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin
Tools is at C:\tools\cuda

Is the %PATH% setting just the location as listed above?

What is the environment variable name and value for each setting?

Or what should the four path entries be.?
All help is appreciated