archive-v2.3.6

From upstream.

X-AnyLabeling v2.3.6

Not all versions of X-AnyLabeling are built as installer/portable binaries for GPU acceleration.

Pushing dependency management away from installer/portable binaries towards dist/OS or worse manual source code compilation, is not sustainable. Any non-essential functionality not available through installer/portable binaries will NOT be supported. If existing documentation is sufficient to build a newer installer of any precompiled binaries possibly needed, that will be preferable to a project scripting compilation on the dist/OS itself.

Long-term, software toolchains will be replaced by portable standalone codebases (eg. reimplementations of AI software in rust or similar), which will serve both automatic (eg. OCR, depth estimation, object detection) as well as non-automatic (eg. training data labeling) uses, and will function as a standalone binary for all relevant platforms (eg. Linux, ubdist/OS, Cygwin/MSW) through only simple pipes, CLI parameters, etc.

'Note: The GPU version mentioned above is compiled based on CUDA 11.6. Different versions may result in issues with xxx functionality. It is recommended to consider running the application using the source code, especially if compatibility issues arise.'
It is potentially acceptable to limit the use of this software only to GPUs and driver versions with the correct, matching version of this 'CUDA' software.