/sdg/

Old Bread

>user Guides
voldy guitard: rentry.org/voldy
inpainting: rentry.org/drfar
textual inversion: rentry.org/aikgx
AMD (docker): rentry.org/sdamd
AMD (onnx): rentry.org/ayymd-stable-diffustion-v1_4-guide
CPU: litechan.org/ran/67

>Web UIs
(Recommended) voldy: github.com/AUTOMATIC1111/stable-diffusion-webui
hlky fork: github.com/sd-webui/stable-diffusion-webui

>Krita Plug-ins
github.com/sddebz/stable-diffusion-krita-plugin
github.com/nousr/koi

>Colabs
voldy: colab.research.google.com/drive/1Iy-xW9t1-OQWhb0hNxueGij8phCyluOh
hlky: colab.research.google.com/drive/1AfAmwLMd_Vx33O9IwY2TmO9wKZ8ABRRa

>Paperspace
github.com/Engineer-of-Stuff/stable-diffusion-paperspace/

>Usage Info
Any Forums wiki: wiki.installgentoo.com/wiki/Stable_Diffusion
SD Akashic Records: github.com/Maks-s/sd-akashic

>Tools & Resources
big index: pharmapsychotic.com/tools.html
artist styles: sgreens.notion.site/sgreens/4ca6f4e229e24da6845b6d49e6b08ae7
prompt sharing: lexica.art/ | krea.ai/
dataset: rom1504.github.io/clip-retrieval/ | laion-aesthetic.datasette.io/laion-aesthetic-6pls/images
interrogrator: replicate.com/methexis-inc/img2prompt

>SaaS Sites
txt2img: huggingface.co/spaces/stabilityai/stable-diffusion
img2img: huggingface.co/spaces/huggingface/diffuse-the-rest
txt2img, new accounts get 200 gens: beta.dreamstudio.ai
txt2img, no account needed: dezgo.com
txt2img, NSFW focused: pornpen.ai

>Finetuned Models
huggingface.co/hakurei/waifu-diffusion
huggingface.co/naclbit/trinart_stable_diffusion_v2

>Related
/h/ /sdhb/: /vg/ /aids/: /mlp/ /ppp/: Official: discord.gg/stablediffusion
NSFW: discord.gg/2ZhBzDGf8u
Tech support: matrix.to/#/#info:conlaboro.xyz
SD updates: twitter.com/EMostaque

Attached: download.png (512x512, 409.67K)

page 6 nigger, and u forgot the whole title

/sdg/ - Stable Diffusion General

Thank you.

shut the fuck up faggot last op didnt even include the tag

retard

Attached: sweaters.jpg (1920x1536, 2.46M)

dezgo is so slow producing fancy booba. I need this vacation to end so I can get back to my PC

Attached: 492CB8DA-1066-4F2D-9CBD-A890C191D7A5.png (512x512, 423.89K)

retard thread
spam thread
nigger thread

we're out of characters for new links

Attached: download.png (512x640, 269.68K)

i really really really like this image, user

no one noticing how gpu usage drops every time the webui has to update the preview? generation is faster too, 1.53s/it vs 1.23s/it with same settings

Attached: Untitled.png (333x88, 2.68K)

Are you talking about live preview? If you're running with --lowvram or --medvram downtime is expected because it has to shuffle models.

nice OP

Nymphbro, I see in you a kindred spirit

Attached: IrishNymphCollage.png (2046x1316, 3.9M)

Proompt?

That's the tradeoff for having a preview feature.

If you ever queue like hundreds of images overnight then remember to turn off the preview.

Yeah I think I will be turning it off completely. I saw some user the other day claiming the preview improved generation time

On my machine, if I turn off medvram/lowram, it's done in parallel with main processing, no downtime.

Of course it does not improve generation time.

> esoteric computing
classic g

Any new big features since negative prompting?

>he doesn't pray to Lain before each batch so that his lovelies may be blessed

Saving a collection of prompts as a present.

Oh, yeah. I still gotta save some presets, that's what I'll do today.

Attached: 382048857.png (448x768, 421.42K)

so you didn't lose the prompt after all.

Should I enable optional generation of txt files with settings for each image in the queue?
What else would you like to see in pygui?

what?

you know what i mean

no, i don't, that's why i asked
what do you mean?