Comments

If you don't mind me asking, which model are you using?

I just noticed I misspelled Mohrbaher's name in the prompt. And it seems I've been copying it ever since. Shame on me.

When you mention denoising strength, is that referring to the upscale setting or something else?

Ocean3 said:

When you mention denoising strength, is that referring to the upscale setting or something else?

It's the one in highres fix. I'm over 500 commits behind on the webui so it might work differently now. (New versions changed the api which broke my autoprompt scripts and I'm too lazy to fix it.)

antlers_anon said:

It's the one in highres fix. I'm over 500 commits behind on the webui so it might work differently now. (New versions changed the api which broke my autoprompt scripts and I'm too lazy to fix it.)

Ah, thanks - I'm not even too sure what that setting does and haven't used it 👌

Ocean3 said:

Ah, thanks - I'm not even too sure what that setting does and haven't used it 👌

It works almost exactly the same as in the img2img tab. I lowered it from the default (0.7 I think) to 0.6 to reduce the amount of mouths and nipples popping in random places (might be placebo though).

antlers_anon said:

It works almost exactly the same as in the img2img tab. I lowered it from the default (0.7 I think) to 0.6 to reduce the amount of mouths and nipples popping in random places (might be placebo though).

Good to know, thanks. Seems to just be built in to the generation I suppose

bro used 80 different models to generate this masterpiece

ForeskinThief said:

bro used 80 different models to generate this masterpiece

Was experimenting with a mass 'anime styled' model mix. I noticed my prompt had a lot of weight on specific things at the time which shifted how the model responded and made me notice how it does with certain things (architecture/food etc.). I've done a few tests since with the random merge and this is one of the results 👌

The Council of Goblin Queens has summoned you. Will you accept their quest?

> Yes
No

redjoe23 said:

The Council of Goblin Queens has summoned you. Will you accept their quest?

> Yes
No

At first I got offended by you calling them goblins. Then I remembered my prompt.
And hell yes, the quest is accept!

Judging by the filename the prompt was something like "peaceful landscape cinematic early morning flat grassy".

sheev_the_senate said:

Judging by the filename the prompt was something like "peaceful landscape cinematic early morning flat grassy".

im guessing grassy field or something, im gonna stop being lazy from now on using picture links and download with metadata instead

When the hands are so good, it's not even bothering it's 6-fingered.

Lyren said:

When the hands are so good, it's not even bothering it's 6-fingered.

Maybe my eyesight is failing me but I don't see an extra finger.

ANJU said:

Maybe my eyesight is failing me but I don't see an extra finger.

Then it ought to be long thumbs that look like index fingers, or it's my eyesight.

Lyren said:

Then it ought to be long thumbs that look like index fingers, or it's my eyesight.

Ah, okay. I see what you mean now.

The hand kinda ruins it but I still like this picture, the face especially.

ANJU said:

The hand kinda ruins it but I still like this picture, the face especially.

oh damn, I didn't even look at the hand.

非常喜欢你这个模型可以分享吗？

I like your model very much. Can you share it 6030dabe

Do you have the hash for AbyssOrangeMix2? And what model is Latte? Image looks great!

kaisu1 said:

Do you have the hash for AbyssOrangeMix2? And what model is Latte? Image looks great!

I used AbyssOrangeMix2_nsfw, and you can check the sha-256 hash here.
https://huggingface.co/WarriorMama777/OrangeMixs/blob/main/Models/AbyssOrangeMix2/AbyssOrangeMix2_nsfw.safetensors

Also you can get Latte in this repo.
https://huggingface.co/andite/yohan-diffusion

@mellohi I fixed your tags, it seems like you've pasted the prompt into the tag field by accident. Please add self_upload to the tags if you generated this image.
Also, you'll get better colors if you add a custom .vae (looks like you're missing one, but I might be wrong).

lovely, good work

https://www.pixiv.net/artworks/104550289

I feel a strange compulsion to pet this ice cream...

I couldn't see it at first, but squinting from a distance worked.

I found a extremly similar post, the face and veins are different.
post #16049

Could you share the model and hypernetwork you use please?
I want to generate images like that too :)

Judging by the desaturated colors you don't have a VAE (variational autoencoder) loaded. I've had the same issue in the past too. You should download one of those and try generating with it selected - the colors look much better that way. vae-ft-mse-840000-ema-pruned is a decent choice, but for anime style images the one from Anything-v3 or NovelAI would probably work better. If you choose either of those two latter ones you'll need to launch automatic-1111 with the --no-half-vae argument to avoid occasional completely black images.

sheev_the_senate said:

Judging by the desaturated colors you don't have a VAE (variational autoencoder) loaded. I've had the same issue in the past too. You should download one of those and try generating with it selected - the colors look much better that way. vae-ft-mse-840000-ema-pruned is a decent choice, but for anime style images the one from Anything-v3 or NovelAI would probably work better. If you choose either of those two latter ones you'll need to launch automatic-1111 with the --no-half-vae argument to avoid occasional completely black images.

Thank you for sharing your experience! Now I know how it works 🙂