I upvoted this one although I can see how it would be challenging to implement with Venice’s ‘everything local’ approach, as a LoRA model can be several 100MB, that would be challenging to store and upload every time (what if you want to use more than one LoRA?), and probably will uncover lots of other snags like possible issues with samplers, schedulers, VAE, etc.
Log in to comment and vote
Comments1
Bronze Meteor
Jul 19, 2025
I upvoted this one although I can see how it would be challenging to implement with Venice’s ‘everything local’ approach, as a LoRA model can be several 100MB, that would be challenging to store and upload every time (what if you want to use more than one LoRA?), and probably will uncover lots of other snags like possible issues with samplers, schedulers, VAE, etc.