Ocpapi - high number of predictions

I have a dataset of candidate oxide/sulfide/nitride catalysts of which I’d like to calculate adsorption energies of different intermediates. I’d like to do this with ocp api. Is it possible to accelerate this process by parallelization or using GPU maybe?

The ocpapi currently doesn’t support custom materials + adsorbate combinations. That feature should be coming in the future. As far as GPU usage - calculations submitted through the api - GitHub - Open-Catalyst-Project/ocpapi is run on machines hosted on our end and are run in parallel. So you would simply need to just submit multiple jobs and it will parallelize automatically.

Let me know if you have any other questions to help you get started.

Thanks for your reply! I could get data in multiple runs as you described.