[Feature request] Will cake be able to train model in the future?

[fish4terrisa-MSDSM]

Cake is very useful for me , as it can make full use of all my devices and share the memory(which allows me to run larger models)
However, it seems that cake currently cannot be use to train LoRA models...
Will we add this feature in the future? That's even more useful(Training uses much more resources)
As cake depends on candle, there exists candle-lora which allows people to use candle to train models.

(And, to save as much memory as possible, wilk cake add support for gguf and 4bit models(QLoRA)? )

[evilsocket]

I'm afraid that the delay added by distributing the network, while acceptable for inference, would make training impractically slow.