GPU utilisation

Dear Robert

Again thank you for help, even with the addition of import ij.IJ; and, "32-bit", ""); we get the same error

INFO: Quadro RTX 8000
ERROR: IllegalArgumentException at line 38: Couldn't instantiate converter found from class ij.CompositeImage to class net.haesleinhuepf.clij.clearcl.ClearCLBuffer

ERROR: net.haesleinhuepf.clij.converters.FallBackCLIJConverterService.getConverter(

We are dealing with large 5 plex IF files trying to run a QuPath classifier script on the whole slide , even with current CPU Power 20C Intel Xeon 6148 2.4 and 256GB RAM its going to take 72 days hence trying to do with GPU.
However I’m not really sure if this going to be possible

Any other suggestion would be welcome


1 Like

Can you share an example image so that we can reproduce the issue?

Thanks :slight_smile:

Thanks @haesleinhuepf here it is

1 Like

Hey @MustafaElshani,

great, thanks for the example image! Would you mind trying again with the updated clupath? You may have to delete the old plugin from the folder C:/Users/<username>/QuPath/extensions (not sure if this is necessary).

The script remains unchanged. After pushing the image to GPU memory, there should be an image stack in your GPU which has three slices (red, green and blue). You can check that by calling


Let me know if it works!


1 Like

Hi Robert
The script ran without errors, I run on the whole slide, the longest wait was, I guess when it is preparing to send the image to GPU, otherwise the actually computation was instantaneous, it’s awesome

Just need to figure out if we can run some QuPath things such cell detection, object classification etc but I do realize this may be more difficult then we original thought



Hey @MustafaElshani,

thanks for testing!

Yes, you basically can go the clij way now. Check out the tutorials and let me know if you need support!