"There are no longer any instances available with the requested specifications."

I've been getting this error a lot lately when trying to deploy a pod even when the GPU is listed as being available. This persists after multiple page refreshes. A feature request is to help with finding an available instance either by giving feedback with the error on the resource that's not matching the availability list (e.g. drive storage) or to allow filtering by more parameters so I can see what's actually available and source a machine that meets my needs.