sd-webui-modelscope-text2video VS lora

Working on it https://github.com/deforum-art/sd-webui-modelscope-text2video/pull/96 (also, will rename the repo to just sd-webui-text2video after that)

They're linked from the main project's REDME.md under the "Where to get the weights" heading. (https://github.com/deforum-art/sd-webui-modelscope-text2video)

Workflow: Text to video AUTO1111 extension https://github.com/deforum-art/sd-webui-modelscope-text2video

Please, give it a try and leave your feedback. Soon fine-tuned models are planned to be usable in the Auto1111 plugin https://github.com/deforum-art/sd-webui-modelscope-text2video/issues/48 as well

Meanwhile, if you have good training vids, it'd be nice to collect them somewhere for the future training, like inside the extension repo's Discussions https://github.com/deforum-art/sd-webui-modelscope-text2video/discussions

Workflow: Text to video AUTO1111 extension https://github.com/deforum-art/sd-webui-modelscope-text2video

Diffusion unet has an "extended" version nowadays that applies to the resnet part as well as the cross-attention: https://github.com/cloneofsimo/lora

Absolutely. But that doesn't matter because you only have to train it at scale, once. There are papers released already that show it's possible to update weights in small sections. You won't have to wait for the next monolithic LLM to drop to get up to date information. It will start to learn in bits and pieces.

No, it's a deep learning thing

Low Rank Adaptation of Large Language Models.

Recently, I have seen the LoRA technique (Low-Rank Adaptation of Large Language Models) as a popular method for fine-tuning LLMs and other models.

Enter LoRA, a method proposed for adapting pre-trained models to specific tasks[2]. By freezing pre-trained model weights and injecting trainable rank decomposition matrices into the transformer architecture, LoRA can reduce the number of trainable parameters and the GPU memory requirement, making the adaptation of LLMs for downstream tasks more feasible.

Open-source LLM projects have largely solved this using Low-Rank Adaptation of Large Language Models (LoRA): https://arxiv.org/abs/2106.09685

Apparently an RTX 4090 running overnight is sufficient to produce a fine-tuned model that can spit out new Harry Potter stories, or whatever...

Alpaca was trained for $600 ($100 for the smaller model) and offers outputs competitive with ChatGTP. https://arxiv.org/abs/2106.09685