How to install TensorFlow GPU on Windows 10 (2017)
Aug 29, 2017 · 2 min read
Prerequisites:
- Windows 10
- Python 3.5
- GPU that supports CUDA 3.0 or higher. (See: Cuda-GPUs)
Step 1: Install Visual Studio Community 2015
- Join the Microsoft Dev Essentials program. (Don’t worry, it’s free)
- Download and Install Visual Studio Community 2015 with Update 3. It might take awhile.

Step 2: Install CUDA 8.0
- Download and Install CUDA Toolkit 8.0.

Step 3: Install cuDNN v6.0 for CUDA 6.0
- Join the NVIDIA Developer Program.
- Download cuDNN v6.0 for CUDA 8.0 for Windows 10.
- Unzip the contents of cuDNN into the CUDA installation directory. (Default: C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v8.0)

Step 4: Install Tensorflow-GPU
Install Tensorflow GPU via pip using:
pip install --upgrade tensorflow-gpuStep 5: Check installation
>>> import tensorflow as tf
>>>
>>> hello = tf.constant('Hello, Aaron!')
>>> sess = tf.Session()
>>> print(sess.run(hello))
b'Hello, Aaron!'Step 6: Rethink your choice of breakfast.
