Tensorflow 2.15 in Windows 11 GPU doesnt work

I’m having problems running Tensorflow on Windows 11

Python 3.11.8
GPU Nvidia 3080
Tensorflow 2.15.0

Tensorflow does not detect the gpu, according to what I was reading Tensorflow has no more gpu support on windows. But I don’t understand why I can’t install 2.10, I can’t find it.

pip install tensorflow==2.10

It returns an error

ERROR: Could not find a version that satisfies the requirement tensorflow==2.10 (from versions: 2.12.0rc0, 2.12.0rc1, 2.12. 0, 2.12.1, 2.13.0rc0, 2.13.0rc1, 2.13.0rc2, 2.13.0, 2.13.1, 2.14.0rc0, 2.14.0rc1, 2.14.0, 2.14.1, 2.15.0rc0, 2.15.0rc1, 2.15.0)
ERROR: No matching distribution found for tensorflow==2.10

Linux is not an option, neither is WSL, we already have the whole project standardized on Windows.

1 Like

Same here. Tensorflow is very difficult to install, I swear

Hi @Franco_Antuna & @Hans-MySql_Raphanels

Welcome to the TensorFlow Forum!

This error is caused because

GPU support on native-Windows is only available for 2.10 or earlier versions, starting in TF 2.11, CUDA build is not supported for Windows. For using TensorFlow GPU on Windows, you will need to build/install TensorFlow in WSL2 or use tensorflow-cpu with TensorFlow-DirectML-Plugin

And TensorFlow 2.10 is compatible with Python 3.7-3.10 as mentioned in this tested build configuration. Which means, installing the TensorFlow 2.10 with Python 3.11 raised the above incompatible version error.

Please try again by downgrading the Python version to 3.7 - 3.10 and install the TensorFlow 2.10 with compatible CUDA and cuDNN by following this TF install official link to have GPU support enabled in your WinOS. Thank you.

WSL worked for me with 2.16.1

  1. Install WSL

  2. pip3 install tensorflow[and-cuda]

import tensorflow as tf

print("Tensorflow version: ", tf.version)

print("Num GPUs Available: ", len(tf.config.list_physical_devices(‘GPU’)))

On Windows guys use WSL2