ClojureCL looks quite awesome, it looks like it takes all of the boiler plate out in comparison to C but I am very interested in the overhead. I tried aparapi for my bachelor thesis and for larger problems it seemed to have quite the overhead that the normal kernel did not have (several years ago). After watching your talk on youtube [1], I saw you compared the GPU implementation to the JVM and other CPU implementations in Clojure. Do you have any numbers for the same kernel being called from a native C openCL application?

[1] https://www.youtube.com/watch?v=bEOOYbscyTs

For equivalent code, you will not experience any noticeable difference compared to C. It is nothing like Aparapi.

The source repository for the lib(s):

https://github.com/uncomplicate/clojurecl

https://github.com/uncomplicate

Part 1: Hello CUDA at http://dragan.rocks/articles/18/Interactive-GPU-Programming-...

PyOpenCL provides a NumPy like array, usable from IPython or Jupyter,

https://documen.tician.de/pyopencl/array.html

It doesn't support the more complex NumPy APIs, but it's a good starting point.

The title should state that it's all in Clojure...

This was the introduction to OpenCL and GPU programming that I was looking for. Great read.

Is it possible to share contexts with OpenGL for visualization ?

Yes. OpenCL supports that. However, I am using it for computation and Machine Learning algorithms, so I haven't tried that myself and can't give you detailed pointers.