Text-to-LoRA: Instant Transformer Adaptionhttps://t.co/DNy6gzwQKN

Generative models can produce text, images, video. They should also be able to generate models! Here, we trained a Hypernetwork to generate new task-specific LoRAs by simply describing the task as a text prompt. https://t.co/QkSaBrZU22

— hardmaru (@hardmaru) 12 juin 2025

Text-to-LoRA: Instant Transformer Adaption https://
arxiv.org/abs/2506.06105 Generative models can produce text, images, video. They should also be able to generate models! Here, we trained a Hypernetwork to generate new task-specific LoRAs by simply describing the task as a text prompt.

→ View original post on X — @hardmaru