Tensorflow – Which TensorFlow and CUDA version combinations are compatible

compatibilitycudacudnntensorflowversion

I have noticed that some newer TensorFlow versions are incompatible with older CUDA and cuDNN versions. Does an overview of the compatible versions or even a list of officially tested combinations exist? I can't find it in the TensorFlow documentation.

Best Answer

TL;DR) See this table: https://www.tensorflow.org/install/source#gpu

Generally:

Check the CUDA version:

cat /usr/local/cuda/version.txt

and cuDNN version:

grep CUDNN_MAJOR -A 2 /usr/local/cuda/include/cudnn.h

and install a combination as given below in the images or here.

The following images and the link provide an overview of the officially supported/tested combinations of CUDA and TensorFlow on Linux, macOS and Windows:

Minor configurations:

Since the given specifications below in some cases might be too broad, here is one specific configuration that works:

tensorflow-gpu==1.12.0
cuda==9.0
cuDNN==7.1.4

The corresponding cudnn can be downloaded here.

Tested build configurations

Please refer to https://www.tensorflow.org/install/source#gpu for a up-to-date compatibility chart (for official TF wheels).

(figures updated May 20, 2020)

Linux GPU

Linux CPU

macOS GPU

macOS CPU

Windows GPU

Windows CPU

Updated as of Dec 5 2020: For the updated information please refer Link for Linux and Link for Windows.

Related Solutions

How to get the CUDA version

As Jared mentions in a comment, from the command line:

nvcc --version

(or /usr/local/cuda/bin/nvcc --version) gives the CUDA compiler version (which matches the toolkit version).

From application code, you can query the runtime API version with

cudaRuntimeGetVersion()

or the driver API version with

cudaDriverGetVersion()

As Daniel points out, deviceQuery is an SDK sample app that queries the above, along with device capabilities.

As others note, you can also check the contents of the version.txt using (e.g., on Mac or Linux)

cat /usr/local/cuda/version.txt

However, if there is another version of the CUDA toolkit installed other than the one symlinked from /usr/local/cuda, this may report an inaccurate version if another version is earlier in your PATH than the above, so use with caution.

Selenium – Which ChromeDriver version is compatible with which Chrome Browser version

Compatibility matrix

Here is a chart of the compatibility between chromedriver and chrome. This information can be found at the Chromedriver downloads page.

chromedriver    chrome
2.46            71-73
2.45            70-72
2.44            69-71
2.43            69-71
2.42            68-70
2.41            67-69
2.40            66-68
2.39            66-68
2.38            65-67
2.37            64-66
2.36            63-65
2.35            62-64
2.34            61-63
2.33            60-62
---------------------
2.28            57+
2.25            54+
2.24            53+
2.22            51+
2.19            44+
2.15            42+

After 2.46, the ChromeDriver major version matches Chrome

chromedriver    chrome
76.0.3809.68    76
75.0.3770.140   75
74.0.3729.6     74
73.0.3683.68    73

It seems compatibility is only guaranteed within that revision.

If you need to run chromedriver across multiple versions of chrome for some reason, well, plug the latest version number of chrome you're using into the Chromedriver version selection guide, then hope for the best. Actual compatibility will depend on the exact versions involved and what features you're using.

All versions are not cross-compatible.

For example, we had a bug today where chromedriver 2.33 was trying to run this on Chrome 65:

((ChromeDriver) driver).findElement(By.id("firstName")).sendKeys("hello")

Due to the navigation changes in Chrome 63, updated in Chromedriver 2.34, we got back

unknown error: call function result missing 'value'

Updating to Chromedriver 2.37 fixed the issue.