invoke-ai
Add Convert Model Endpoint
Updates to add a (working) Convert model endpoint.
Could do with some enhancements to add a 202, but will need an adult to help with what else that would entail. Socket emit once complete?
One of the problems with the current WebUI is that there is no feedback during the 30s it takes to covert a model. However there are defined steps in conversion as each part of the model is converted. Works it be helpful to add yield or callbacks to convert so that progress messages can be provided to the user?
There is minimal feedback. The status text at the top changes to Converting Model and the call to action buttons all get disabled or show a loading indicator. But if there was a callback with better options and status messages, it'd be nice to have it implemented.
@Kyle0654 - Any thoughts on what this conversion process would entail?
In a world where this is controlled by API, unclear whether this would lock up generation, or begin processing and then report back later, and what that pattern looks like. Maybe some input from @psychedelicious or @blessedcoolant on that front - (i.e., is it all the same? would that be handled automatically b/c the conversion process would be getting initiated, or would that need to get patched up elsewhere)
It's not a node, so it wouldn't necessarily lock anything up. I don't know how you'd want to signal on progress though. Mostly how the client would know to listen for signals, how other clients would behave, etc.
