/sdg/ - Stable Diffusion General

Old bread: >Local install
GPU: rentry.org/voldy
CPU only: rentry.org/cputard

>AMD GPU guides
Docker: rentry.org/sdamd
Onnx: rentry.org/ayymd-stable-diffustion-v1_4-guide
Docker on Arch: rentry.org/amddockerarch

>Cloud hosted install
Colab: colab.research.google.com/drive/1Iy-xW9t1-OQWhb0hNxueGij8phCyluOh
Paperspace: rentry.org/865dy

>SaaS Sites
txt2img: huggingface.co/spaces/stabilityai/stable-diffusion | beta.dreamstudio.ai | dezgo.com | (NSFW) pornpen.ai
img2img: huggingface.co/spaces/huggingface/diffuse-the-rest

>Other guides
Inpainting/Outpainting: rentry.org/drfar
Upscaling images: rentry.org/sdupscale | rentry.org/b7vcb
Textual inversion: rentry.org/textard
Training: rentry.org/informal-training-guide

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

>Krita Plug-ins
github.com/sddebz/stable-diffusion-krita-plugin
flyingdog.de/sd/en

>Tools & Resources
Big index: pharmapsychotic.com/tools.html
Artist styles: sgreens.notion.site/sgreens/4ca6f4e229e24da6845b6d49e6b08ae7 | rentry.org/artists_sd-v1-4
Prompt sharing: lexica.art/ | krea.ai/
Dataset: rom1504.github.io/clip-retrieval/ | laion-aesthetic.datasette.io/laion-aesthetic-6pls/images

>Finetuned Models
huggingface.co/hakurei/waifu-diffusion
huggingface.co/naclbit/trinart_stable_diffusion_v2
Std 1.4 + Waifu 1.2 (linear interpolation "mix"):: mega.nz/file/IwxQTQCb#tsfYZeHrd6U3fg5QCmBazEL-OjOmy1gmhvKAtp1YWFs

>Related
/h/ /sdhb/: /vg/ /aids/: /mlp/ /ppp/: Official: discord.gg/stablediffusion
NSFW: discord.gg/2ZhBzDGf8u
Training: discord.gg/xVsyrmhQWS

Stability.ai updates: twitter.com/EMostaque

Attached: fsdjt.png (512x512, 657.17K)

Other urls found in this thread:

github.com/basujindal/stable-diffusion
twitter.com/SFWRedditVideos

Posting something interesting before the anime tit posters get in

Attached: 1736233-0.png (576x576, 507.23K)

Attached: 00004-2789052132.png (512x512, 264.62K)

Attached: 00063-2313156918-Albert Lynch, Sad Blonde Girl, Armor, Flowing Hair.png (512x512, 340.99K)

Attached: 1663169796648-2.png (512x1024, 624.19K)

is this from that mspaint image that was posted days ago? I'm pretty sure it is, cool how it evolved over time

damn I really hope they make an AI/SD board. those threads reach their limit fast

Attached: 00042-4092010140-realistic landscape, Czeslaw Znamierowski, intricate details, midday, green forest, wide angle-gigapixel-standard-scale-4_00x.jpg (2816x2048, 1.58M)

what did you use to get the arms up and the hands behind the head?

No its original

There'll be no accusations, just friendly crustaceans, under the sea

Attached: 00151-560-water_on_lens,_rooftop_scene,_((underwater)),_((cityscape)),_skyscrapers,_submarines,_fish,_divers,_((bubbles)),_glass_domed_bui.png (512x512, 453.39K)

Attached: 4.png (448x576, 364.66K)

please do, im waiting for my new ram sticks and crayion isnt doing much for me.
Im trying to use ai for concept art and being able to paint in areas with sd will be really excting to try out

Attached: last ned (25).jpg (256x256, 17.39K)

Durgasoft Boss

Attached: bhima.jpg (1536x2304, 438.38K)

LOL nevermind then

Can this be used with AUTOMATIC1111? github.com/basujindal/stable-diffusion

Attached: 34343.png (512x768, 522.78K)

Attached: 5zrhr.png (512x512, 328.53K)

Prompt pleaseeeeee

Attached: download (11).png (512x512, 301.02K)

>After 2 prompts run out of memory.
Feels bad man.

It seems really good at kino cyberpunk stuff

((spread armpit)) with more or less ()

Attached: download (3).png (512x1024, 701.58K)

Which prompts?

Remember to eat your greens

Attached: Ireland4.png (512x658, 341.42K)

Attached: 00239-1913090782-Cute_blonde_loli,_realistic_proportions,_highly_detailed,_smooth,_sharp_focus,_8k,_ray_tracing,_digital_painting,_concept_art_il.png (512x512, 298.43K)

Noice

Catholic++

Attached: 00454-90101-Ein_Kreuzritter_aus_dem_15._Jahrhundert,_er_hat_sehr_langes_Haar,_trägt_einen_Brustpanzer_aus_dem_15._Jahrhundert,_ein_Kreuz_auf.png (512x768, 683.15K)

Attached: 00097-470413476-volcano erupting in a snowy landscape and lava flowing, detailed landscape, by Albert Bierstadt.png (768x448, 461.58K)

thanks for the update, the colab works again

She looks like she stinks good

maybe try English prompts if you want more accurate results. I don't think the AI is trained for German

you need doggettx mods along with --medvram if you want to do really big images

At a certain point the prompts themselves become pointless and then it's playing with image to image

Attached: 00116.png (1024x1536, 1.65M)

Attached: 00004-~1.png (512x512, 324.53K)

I know, I was testing

Ok but how to you get to show the whole body for example?

Attached: 00003-30_k_euler_a_3698467344.png (512x512, 321.54K)

Why does mine look like this?

Attached: Screenshot 2022-09-14 at 09-46-17 Stable Diffusion.png (1482x162, 26.3K)

Attached: 00120-.jpg (1536x2304, 485.48K)

Nice

Oh yeah do this
Full Body Shot
Add emphasis as desired
This made the biggest difference
Good luck

Didn't Auto1111/Voldy include Dogg's changes with the split attention command line argument? Or was that for something else?

this shit is ridiculously addicting. I spent pretty much all day prooooompting. This morning, I'm back at it.

What we really need is a better combination of the SD updated model (I've heard v1.5 is online or something but not available for download yet?) mixed with a lot of erotic photography but not necessarily porn. The ability to create high fidelity erotic images would be a game changer - seems like right now you more or less have to use the lewd model or the waifu model.

I think the program could also really use a model selector, where you can have a folder of models and then just select from a dropdown which model to use for generation.

Attached: 00295-16351680851761-Artstation, artgerm, ((alphonse mucha)), cgsociety. (Ice queen Elsa from Frozen). (nude small breasts). visible (nipples). Midri.png (512x768, 692.8K)

>paganism
No thanks

ah ok

Attached: 00135-124569-holy crusader from the 13th century standing with cross in hand, ((lightning bolt in the background)), by Winslow Homer.png (512x704, 700.76K)

Yeah its great

Attached: 00000-~1.png (512x512, 395K)