Ybzhang2001

InstantStyle appears to perform poorly on Playground-v2.5-1024px-aesthetic.

2

Playground-v2.5-1024px-aesthetic is a stronger basic T2I model which follows the same architecture as Stable Diffusion XL. (https://huggingface.co/playgroundai/playground-v2.5-1024px-aesthetic) I try to employ InstantStyle on the model but the result perform poorly....

Hi, thanks for the open-sourced project! I try to use my own mesh (without texture) as the initialization of stage 3 and hope to optimize the corresponding SuGaR using default...

Question about BS-Objaverse.

2

Hi，thanks for your wonderful work. I find that the BS Objaverse dataset has some data with scores of 6, but the definition in the paper only includes scores of 1-5....

Question about the training cost.

4

Hi, thanks for your excellent work! I would like to know more about the training cost of these three models. At the same time, I ran the ModelScope script on...

[Bug] OOM when fine-tuning WAN 2.1-14B on 8x H200 (batch=1, V1 script)

9

### Describe the bug Using the V1 script to fine-tune WAN 2.1-14B on 8×H200 GPUs with a batch size of 1 and input data of shape 512×512×81 results in an...

[Feature] Thanks for adding VSA support - question about compatibility with larger models

2

### Motivation Thank you for adding VSA support for Wan2.1 T2V 1.3B — the ```finetune_v1_VSA.sh``` script is very helpful! I wonder if the current script also supports the Wan2.1 T2V...