watch_movie

This is a fine tuning stable diffusion model, baseed on chilloutmix, I try to make the texture of the picture as much as possible; This check point contains all the model weights (text, Unet, vae). You can try substituting, which might get better results. At the same time, I hope I can get more opinions to adjust the model.

Attention: Due to an update to the civitai website, the model file version selection has been moved to the top of the page. There may also be problems that cannot be displayed. Please note in the comments section if you cannot see the version selection.

Usage

Load the model, which has no fixed parameter Settings, you can use any sampler; Of course, I have provided some examples and supporting parameters for your reference.

Suggested negative prompt:

sketches, (worst quality:2), (low quality:2), (normal quality:2), lowres, normal quality, ((grayscale)), skin spots, skin blemishes, bad anatomy, ((monochrome)), (((extra legs))), ((grayscale)),DeepNegative, tilted head, lowres, bad a natomy, bad hands, text, error, fewer digits, cropped, worstquality, low quality, bad legs, fused fingers,too many fingers,long neck,cross-eyed,mutated hands,polar lowres,bad body,bad proportions,gross proportions,missing fingers,missing arms,missing legs,extra digit , extra arms, extra leg, extra foot

CFG：

This number needs to be slightly larger: 9-12.

Versions

2023-04-04 update Pruning: I uploaded the pruning model in the latest version, and the model capacity decreased significantly.

note:

The model is still a work in progress, and I hope users will give me more feedback.