sd_hassan

My prompt wildcard (full prompts)

Attached is a wildcard of all the prompts I use to generate images. Somedays I just run this wildcard on batches and create nice random outputs. 

Just go to your WebUI, install the wildcard extension and drop this prompts.txt file into your folder:

stable-diffusion-webui\extensions\stable-diffusion-webui-wildcards\wildcards

Then in your prompt itself, use the name of the text file with two underscores before and after.&n...

Working on a new finetune

for around 300k steps if all goes well.

It will be finetuning hassansblend with photorealistic content related to female anatomy, sex positions , male anatomy, general portrait stuff
Scraped from places like the top NSFW subreddits/nsfw websites.

It's a trial finetune and if it goes well I will scale it up larger. If it works well, I will start to try to finetune SD2.0 at a larger scale.

It's all done on the cloud so I appreciate all the support from you folks ...

D.B embedding

higher steps for supershot

D.B embedding

11,500steps for boosted

H.P higher step embedding

trained on 30 images for various steps depending on tier. Trained on sd1.5

example: a lovely photo of  stunning haydenpanettiere woman with a smile on her face, wearing a cheerleader outfit, perfect symmetrical face, toned abs


This will go up to 20k steps so i will add the rest as they complete training

H.P new embedding

trained on 30 images for various steps depending on tier. Trained on sd1.5


example: a lovely photo of  stunning haydenpanettiere woman with a smile on her face, wearing a cheerleader outfit, perfect symmetrical face, toned abs

SFW womens portraits higher steps

SFW womens portraits to improve photo consistency and skin texture along with general composure. over 1k image data used in training.

-

-31k steps for this tier so far, training up to 200k steps overall

I'd recommend low strengths, 0.2 to 0.4 , anything over may have too much effect

New hypernetwork - womens portraits

SFW womens portraits to improve photo consistency and skin texture along with general composure. over 1k image data used in training.

-

-16k steps for this tier, training up to 200k steps overall

Hypernetwork - Female Poses

Hypernetwork trained on female poses. 200 images - to improve overall consistency of females in images.

.

.

.

.

Trained a new hypernetwork up to 130k steps on women poses to work with Hassans models.
-137k steps for this tier

I'd recommend low strengths, 0.2 to 0.4 , anything over may have too much effect

New hypernetwork - womens portraits

SFW womens portraits to improve photo consistency and skin texture along with general composure. over 1k image data used in training.

-

-5k steps for this tier, training up to 200k steps overall

New Hypernetwork - female poses

Hypernetwork trained on female poses. 200 images - to improve overall consistency of females in images.

.

.

.

.

Trained a new hypernetwork up to 130k steps on women poses to work with Hassans models. This tier gains access to 5