Super protocol Install miniconda https://conda.io/miniconda.html Install Latest nvidia driver current driver: 417.35 Install CUDA Toolkit v9.0 Get the Patch if any During installation, # Select Advanced # Select CUDA # Deselect Visual Studio Integration Install cuDNN for tghe 9.0 version of CUDA Download Zip file (You need to register) https://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#installwindows Unzip it into `C:\tools\` You just need to **copy **the //contents// of the `cuda` folder into the CUDA installation which should be `C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v9.0` Add Environment variables # C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v9.0\bin # C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v9.0\extras\CUPTI\libx64 # C:\tools\cuda\bin That last one might not be necessary Open Anaconda Prompt conda update conda conda update python conda update --all Make a directory for all the tensorflow projects somewhere mkdir tensorflow-projects cd tensorflow-projects Create the environement conda create --name tf-gpu python=3.6 conda activate tf-gpu pip install tensorflow-gpu