We couldn't agree more! SLMs fine-tuned with LoRA for specific tasks is a much more resource (and cost) efficient approach! Love your video walkthrough, it's very thorough. What's your preferred way of serving multiple LoRAs?

→ View original post on X — @predibase,