trouble-maker007

@vaibhavad thanks for the response, I just wonder after llm2vec, llm model a better text encoder than t5 in sd training

@Artanic30 Thanks for the quick response. And did you compare with flant5-xxl as the text encoder

@hyw-dev I think use this one to generate svg file should be a better choice: https://github.com/ivanpuhachov/line-drawing-vectorization-polyvector-flow, but this code depend on a library [Gurobi](https://github.com/ivanpuhachov/line-drawing-vectorization-polyvector-flow#gurobi), that is not open source

@limuloo thanks for quick response, I should take a try

@chaofengc support sdxl weights？

@LinB203 Thank you for your reply. I apologize for only noticing the checkpoint link in the report.

first frame for image

@Robertwyq May I ask your fine-tuning method for image to video is the same as described in the paper?

@yzy-thu @tengjiayan20 But I found that after training for a while with the same strategy as SVD, the dynamic changes are similar to SVD, both with relatively small amplitudes of...