-  [WT]  [PS]  [Home] [Manage]

Posting mode: Reply
  1.   (reply to 43)
  2.   Help
  3. (for post and file deletion)
/ai/ - AI-Generated Artwork & Text

Share your AI-generated waifu and your fantasy worlds, text-based outputs, models, tools, how-to's, and more.

NOTE 1: Realistic loli/shota falls under the "Absolutely no child pornography" rule. If the average boomer might think they're real, they're real. Loli/shota is fine, but they better look less realistic than that bitch from Alita: Battle Angel.

NOTE 2: Yes, we know language models can teach you how to do illegal shit. No, you can't post it here.

  • Supported file types are: GIF, JPG, PNG, SVG, TXT, WEBM
  • Maximum file size allowed is 4000 KB.
  • Images greater than 200x200 pixels will be thumbnailed.
  • Currently 232 unique user posts. View catalog

  • Blotter updated: 2018-08-24 Show/Hide Show All

We are in the process of fixing long-standing bugs with the thread reader. This will probably cause more bugs for a short period of time. Buckle up.

Movies & TV 24/7 via Channel7: Web Player, .m3u file. Music via Radio7: Web Player, .m3u file.

WebM is now available sitewide! Please check this thread for more info.

AnonGPT 23/05/05(Fri)11:13 No. 43 ID: c284aa

File 168327802230.jpg - (26.53KB , 512x446 , 9007c6672c8106cf9c4818065b2cc25c531844e5_00.jpg )

Could you share a prompt, bo‌ther, for the collective?

Also, vaginal penetration, plx.

☑I am not a robot.

AnonGPT 23/05/05(Fri)12:52 No. 44 ID: d9cb6e

trial and error, just either use these or other ways like, https://dezgo.com/ or https://www.mage.space/ create a login and use pics
anyhow, trial and error

>>> Install <<<

>> Models/Embeddings/Characters/Celebs/fetish content <<

>> Low VRam or Radeon card or render while jobbing? Use the google collab <<
set gdrive location as /content/gdrive/MyDrive/
DL content directly from civitai to Gdrive with the following honk.
!cd /content/gdrive/MyDrive/sd/stable-diffusion-webui/models/; wget https://civitai.com/api/download/models/15640 --content-disposition

>> Faded colors/Low quality/blurry/aliased? Get a VAE <<
Some models don't sneed a VAE

>>>More info<<<

>>> Xformers for cross attention layers (works on Google collab)<<<
go from 960x960 being your resolution limit to 1600x1600 and Reduce legacy rendering honk by 30%

AnonGPT 23/05/06(Sat)06:33 No. 47 ID: 1d93bc

File 168334759414.png - (630.24KB , 768x512 , c21af568ef66235c86818be4a14cb7af.png )

here's one from https://aibooru.online some pics include prompts or metadata, oh and don't forget the negative prompts (what you don't want in pic)
masterpiece, best quality, nsfw, {1girl solo}{{extremely detail}} cg 8k wallpaper, leaning towards watercolor, colorful sketch, {{{{masterpiece}}}}, {{{a lot of tentacles under clothes and cock in thighhighs}}}, {tall supermodel body}, {kawaii face}{{{triple penetration with tentacles}}, {{{tentacle insert breasts cleavage and pussy and ejaculation, cum on full body }}}}, {panties around one leg}, {{{a lot of tentacles under clothes and cock in thighhighs}}}, {{cum and tentacles all over girl}}, depth of field, tiny pink nipple, {{{strongest soft highlight from viewer}}}}, { {{{gorgeous noble blond knight, {{{sex with tentacles to ejaculation}}}, girl's large beasts wrapped in tentacles and cum and ejaculation}}}, {{bound ankles}}, {{show full body}}, { tentacles groping girl full body}, elegant and cool face, {{{strongest highlight, lighten and Diffuse Glow light effect skin}}}, {{ {{{{{{{white tender skin reflects pure white light}}}}}}, {{{{{gorgeous and ribbon and bejeweled style, torn white gorgeous gold lace {tentacle clothes knight armor} and { Lace Rose Gloves} and {Lace long dress} and {Lace Over-the-knee Elf Boots} and {Lace Cape}}, {scattered {{bright blond long hair between eyes}}}, beautiful detailed blue eyes, background is a crowd of rotting zombies and monsters. best quality, illustration, highly detailed, beautiful light, {five fingers}, {strongest soft highlight}, detailed face
lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, nsfw, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, lower, bad anatomy, bad hands, bad feet,clothes,bra, {{yellow skin}}, {{white hair}}, {{yellow light}},{{brown hair}}, bad fingers, {{{elf}}},elf ear

AnonGPT 23/05/07(Sun)03:17 No. 55 ID: c53ccd

File 168342223614.png - (162.08KB , 524x512 , Hugh in 2399.png )

I just want you to know, that I see it's a prompt for tentacle porn.

I am going to try to render it anyway; I need to learn how to do this.

AnonGPT 23/05/07(Sun)11:37 No. 58 ID: 58e36c

File 168345226839.png - (741.08KB , 768x768 , d528c12a43380569a3e1428a96a91be1.png )

>>55 again https://aibooru.online/ use their search

beautiful, masterpiece, best quality, perfect lighting, detailed face, 1girl, solo, <lora:estelleBrightSoraNoKiseki_v1:1.0>, estelle, orange hair, red eyes, twintails, long hair, side bangs, ahoge, bedroom, window, indoors, city background, night, naked, topless, shirtless, medium breasts, <lora:povMissionary_v10:1>, 1boy, large penis, sex, on back, (black stockings:1.1), arms behind head, looking away, pov, pillow, heavy breathing, panting, surprised, open mouth, tongue out, (cum on body:1.3), (cum on breasts:1.3), drooling, trembling, motion lines, wet skin
(easynegative:1.0), (worst quality, low quality:1.2), (bad anatomy:1.4), (realistic:1.1), nose, lips, fat, sad, (inaccurate limb:1.2), extra digit, fewer digits, six fingers, (monochrome:0.95), bad mouth, deformed mouth, poor drawn tongue, bad tongue, blurry eyes, watermark, bad-hands-5, amputee, extra legs, tail, gloves

AnonGPT 23/05/07(Sun)13:17 No. 59 ID: 7c19a9

In the prompts here, anything within < and > such as:

> <lora:estelleBrightSoraNoKiseki_v1:1.0>
> <lora:povMissionary_v10:1>

are referring to LoRA files.

There are also references other files that are less obvious, such as:

> (easynegative:1.0)
> bad-hands-5

These refer to "Textual Inversion" files. Both LoRA and Textual Inversion "steer" your generation subject matter to include, or, in the case of negative versions of them, remove certain aspects of the image.

Textual inversions usually take the place of a long string of prompts (though it does still "cost" prompts to use), whereas LoRA only takes up one prompt at the cost of some negligible additional generation time.

AnonGPT 23/05/14(Sun)06:45 No. 80 ID: 3eaf20

File 168403955030.png - (346.83KB , 512x512 , blob_https___waifus_nemusona_com_6c9cc5a8-4c88-4a4.png )

[Return] [Entire Thread] [Last 50 posts]

Delete post []
Report post