What versions of major Python dependencies are supported?

AI SDK How-To
Product
AI Software Development Kit
Product Version
2.4.0
Language
en-US

The latest version of AI SDK supports the following version of major Python dependencies

Python packages for AI SDK PythonComponent using Python 3.10

  • numpy up to version 2.1.3 (numpy-2.1.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • joblib up to version 1.4.2 (joblib-1.4.2-py3-none-any.whl)

  • pandas up to version 2.2.3 (pandas-2.2.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • scikit-learn up to version 1.5.2 (scikit_learn-1.5.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • opencv-python-headless up to version 4.10.0.84 (opencv_python_headless-4.10.0.84-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • pillow up to version 11.0.0 (pillow-11.0.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • torch up to version 2.5.1+cpu (torch-2.5.1+cpu-cp310-cp310-linux_x86_64.whl)

  • tensorflow up to version 2.18.0 (tensorflow-2.18.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • onnx up to version 1.17.0 (onnx-1.17.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • onnxruntime up to version 1.20.1 (onnxruntime-1.20.1-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl)

  • keras up to version 3.6.0 (keras-3.6.0-py3-none-any.whl)

  • scikit-image up to version 0.24.0 (scikit_image-0.24.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • scipy up to version 1.14.1 (scipy-1.14.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

Python packages for AI SDK PythonComponent using Python 3.11

  • numpy up to version 2.1.3 (numpy-2.1.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • joblib up to version 1.4.2 (joblib-1.4.2-py3-none-any.whl)

  • pandas up to version 2.2.3 (pandas-2.2.3-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • scikit-learn up to version 1.5.2 (scikit_learn-1.5.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • opencv-python-headless up to version 4.10.0.84 (opencv_python_headless-4.10.0.84-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • pillow up to version 11.0.0 (pillow-11.0.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • torch up to version 2.5.1+cpu (torch-2.5.1+cpu-cp311-cp311-linux_x86_64.whl)

  • tensorflow up to version 2.18.0 (tensorflow-2.18.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • onnx up to version 1.17.0 (onnx-1.17.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • onnxruntime up to version 1.20.1 (onnxruntime-1.20.1-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl)

  • keras up to version 3.6.0 (keras-3.6.0-py3-none-any.whl)

  • scikit-image up to version 0.24.0 (scikit_image-0.24.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)

  • scipy up to version 1.14.1 (scipy-1.14.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl)