Tantabus
Filters (Default) SettingsRegisterLogin
Character LoRAs - Want more character variety? Then check This List!

Posts

For more information, see the search syntax documentation. Search results are sorted by creation date.

Search Results

Creative Corner » ADetailer » Post 2

Sunny

I don’t really see the point of an ‘ADetailer’ for cutie marks when you could just use an editing app to add the correct cutie mark manually. I get the idea of using an ADetailer for things like making horse pussy more correct / accurate, but why make a cutie mark-specific ADetailer LORA if the AI is going to keep messing up the cutie marks anyway? I can understand the usefulness of using an ‘ADetailer’ for cutie marks in 3D images, but it doesn’t seem as necessary for 2D ones.
Posted Report
Edited

Creative Corner » ADetailer » Post 1

AIPonyAnon

@derp621
I do the same process as adetailer a lot, except for the automatic masking. The models themselves don’t actually change anything, they just automatically mask areas, which is then used for low-denoise only-masked inpainting.
Posted Report

Creative Corner » ADetailer » Topic Opener

Creative Corner » Text-to-image prompting » Post 25

tyto4tme4l

Something of an artist
@MareStare
Great idea! Img2img and inpainting are invaluable tools and it would be useful to describe them in detail. Things like denoising strength and the difference between “Whole picture” vs “Only masked” for the inpaint area are extremely important here.
Posted Report

Creative Corner » Text-to-image prompting » Post 24

MareStare

DJ HORN3 took the wheel
@Thoryn
I used Photoshop for drawing the scribbles, and then moved the images to Forge UI for inpainting. I’m planning to describe some of my learnings and creative process in a shared guide website. I’ll post about it on tantabus discord and create a forum thread on tantabus when it’s more-or-less ready. I’d like to collect all the tips and tricks and organize them on a convenient medium for beginners study.
Posted Report
Edited

Creative Corner » Text-to-image prompting » Post 23

Thoryn

Latter Liaison
@MareStare
Really cool to see WIP steps like this and have them explained.
And your sketching abilities are way ahead of mine. :p
What program do you use for SD?
Do you handle all the painting in it, or do it elsewhere and move it over to SD?
Posted Report

Creative Corner » Text-to-image prompting » Post 22

MareStare

DJ HORN3 took the wheel
@Thoryn
Getting the composition as you want may be too big of a work for text2img. I recommend you to try inpainting with a colored scribble (coloring is important to make AI get the colors right).
For example, this is how I got Fluttershy inplanted into the scenery of this image:
Yeah, you can tell my drawing skills aren’t that good, but Zoinksnoob nailed Flutty almost immediately after I pasted it there and inpainted that area with a denoising strength of something like 0.7+. Sometimes it takes several iterations of drawing a scribble, then letting inpainting improve the detail, and then improve that more detailed version with some lighter scribble to get things exactly as you want.
Posted Report

Creative Corner » Post Your AI Art! » Post 38

Creative Corner » Text-to-image prompting » Post 21

Thoryn

Latter Liaison
@Sunny
Yes, tried 3 or 4 of classroom LoRAs, and played around adjusting LoRA weights and prompts with them.
Typical problems I had with them, would be thatit insisted on the pony being at the very forefront / focus of the image and taking up half the image, very stretchy pony bodies, wrong size in relation to the room, messed up wall corners and such, pony bodies morphing into the desks, getting desks to look coherent (size, legs, also problem morphing into each other)
It can probably be fixed with extensive inpainting or something, but I should be able to get decent results with just prompting correctly, even without using LoRAs. (Most of my time on this has been spent without them, and I’ve gotten some half-decent results.)
Thinking I should probably look into ControlNet, but am uncertain if it will detect and adjust equine poses correctly.
Posted Report

Creative Corner » Text-to-image prompting » Post 20

Creative Corner » Text-to-image prompting » Post 19

Thoryn

Latter Liaison
Spent hours doing a classroom scene, without even getting the room to where I was satisfied with it, let alone the pony and the pose and interaction within the room.
The tips I received earlier with describing the room in greater detail (floor, ceiling, walls, wall decor, windows..) has proven to be an invaluable advice, but it’s often not enough, at least not the way I’ve done it.
Then after some (ok, many..) hours of fiddling with that, I decided to do another idea I had, which would focus mainly on a basic flank pose and the tail. For a change of scenery, I plopped down a single line describing a bedroom just to have something other than a classroom going on in the image, and it created lots of beautiful bedroom images. (Getting the tail to how I want it seems to be more difficult in this instance compared to the rather basic pose I tried in the classroom scene though.)
Long story short, prompting difficulty can vary greatly just by what location you try to prompt for. Bedrooms, at first glance for me at least, seem to be on the easier side. My main issues with them though, is that the bed is often malformed, and the size isn’t correct ratio to the character.
Posted Report

Creative Corner » Pony Wildcards » Post 2

Lord Waite

Incidentally, if someone was trying to use this with, say, the dynamic prompts extension, you’d save it in the wildcards folder, and type a prompt like:
__pony/header/pony_sfw__,
__pony/mare__ in a forest looking back at you
The “untrained” section, IIRC, was where I was sticking any character names that didn’t seem like they were working properly for me at the time…

Creative Corner » Pony Wildcards » Post 1

Lord Waite

Well, I had this sitting around on my hard drive as ‘pony.json’, but I’m pretty sure it’s something I threw together one day and forgot about…
{
    "pony": {
        "header": {
            "score": [
                "score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up"
            ],
            "pony_sfw": [
                "__pony/header/score__, rating_safe, source_pony"
            ],
            "pony_questionable": [
                "__pony/header/score__, rating_questionable, source_pony"
            ],
            "pony_explicit": [
                "__pony/header/score__, rating_explicit, source_pony"
            ],
            "furry_sfw": [
                "__pony/header/score__, rating_safe, source_furry"
            ],
            "furry_questionable": [
                "__pony/header/score__, rating_questionable, source_furry"
            ],
            "furry_explicit": [
                "__pony/header/score__, rating_explicit, source_furry"
            ],
            "anime_sfw": [
                "__pony/header/score__, rating_safe, source_anime"
            ],
            "anime_questionable": [
                "__pony/header/score__, rating_questionable, source_anime"
            ],
            "anime_explicit": [
                "__pony/header/score__, rating_explicit, source_anime"
            ],
            "cartoon_sfw": [
                "__pony/header/score__, rating_safe, source_cartoon"
            ],
            "cartoon_questionable": [
                "__pony/header/score__, rating_questionable, source_cartoon"
            ],
            "cartoon_explicit": [
                "__pony/header/score__, rating_explicit, source_cartoon"
            ]
        },
        "cmc": [
            "Scootaloo",
            "Sweetie Belle",
            "Apple Bloom",
            "Babs Seed",
            "Gabby"
        ],
        "untrained": {
            "filly": [
                "Zipporwhill",
                "Dinky Doo",
                "Peachy Pie",
                "Sunny Daze",
                "Wind Sprint",
                "Lily Longsocks",
                "filly scout Tag-A-Long",
                "Ruby Pinch",
                "Snowdrop",
                "Nyx"
            ],
            "colt": [
                "Featherweight",
                "Pipsqueak",
                "Snails",
                "Shady Daze",
                "Tender Taps"
            ],
            "mare": [
                "Fleetfoot",
                "Auntie Lofty",
                "Aunty Holiday",
                "Cloud Chaser",
                "Cloud Kicker",
                "Daisy",
                "Flitter",
                "Blossomforth",
                "Raindrops",
                "Twinkleshine",
                "Suri Polomare",
                "Upper Crust",
                "Potion Nova",
                "Jazz Hooves"
    
            ],
            "stallion": [
                "{Flim|Flam}",
                "Argyle Starshine",
                "Bulk Biceps",
                "Caramel",
                "Carrot Cake",
                "Doctor Caballeron",
                "Doctor Hooves",
                "Double Diamond",
                "Filthy Rich",
                "Prince Blueblood",
                "{Fancy Pants|Fancypants}",
                "Gladmane",
                "Hoity Toity",
                "Hondo Flanks",
                "Jet Set",
                "Mudbriar",
                "Bow Hothoof",
                "Trenderhoof",
                "Troubleshoes",
                "{Star Swirl|Starswirl} the Bearded",
                "Zesty Gourmand"
            ]
        },
        "filly": [
            "Scootaloo",
            "Sweetie Belle",
            "Apple Bloom",
            "Babs Seed",
            "Silver Spoon",
            "Diamond Tiara",
            "Cozy Glow",
            "Flurry Heart",
            "Twist",
            "{|Princess }Woona"

        ],
        "colt": [
            "Sandbar",
            "Snips",
            "Button Mash",
            "Rumble"
        ],
        "mane_six": [
            "Applejack",
            "Twilight Sparkle",
            "Pinkie Pie",
            "Rarity",
            "Fluttershy",
            "Rainbow Dash"
        ],
        "mare": [
            "{Bonbon|Sweetie Drops}",
            "Minuette",
            "Derpy{| Doo| Hooves}",
            "{Vinyl Scratch",
            "Octavia Melody",
            "{Golden Harvest|Carrot Top}",
            "{The Great and Powerful Trixie|Trixie Lulamoon",
            "{Aloe Vera|Lotus Blossom} spa",
            "Amethyst Star",
            "Applejack",
            "Berry Punch",
            "Cheerilee",
            "Cherry Berry (yellow mane)",
            "Coco Pommel",
            "Cookie Crumbles",
            "Cup Cake",
            "Lily Valley",
            "{Fleur de Lis|Fleur Dis Lee}",
            "Fluttershy",
            "Granny Smith",
            "Izzy Moonbow",
            "Lemon Hearts",
            "Lightning Dust",
            "Limestone Pie",
            "Marble Pie",
            "Maud Pie",
            "Lyra Heartstrings",
            "Mayor Mare",
            "Misty Brightdawn",
            "Moondancer",
            "Ms. Harshwhinny",
            "Ms. Peachbottom",
            "Night Glider",
            "Nurse Redheart",
            "Opaline Arcana",
            "Pear Butter",
            "Phyllis Cloverleaf",
            "Pinkie Pie",
            "Pipp Petals",
            "Posey Bloom",
            "Princess {Cadence|Cadance}",
            "Princess Celestia",
            "{Princess |}{Luna|Woona}",
            "Queen Chrysalis",
            "Queen Haven",
            "Rainbow Dash",
            "Rarity",
            "Roseluck",
            "Saffron Masala",
            "Sapphire Shores",
            "Sassy Saddles",
            "Somnambula",
            "Spitfire",
            "Spoiled Rich",
            "Starlight Glimmer",
            "Sugar Belle",
            "Sunny Starscout",
            "Sunset Shimmer",
            "Tempest Shadow",
            "Tree Hugger",
            "{Princess |}Twilight Sparkle",
            "Twilight Velvet",
            "Windy Whistles",
            "Zipp Storm",
            "Creamy Heart"

        ],
        "stallion": [
            "Big Macintosh",
            "Alphabittle Blossomforth",
            "Braeburn",
            "Bright Mac",
            "Cheese Sandwich",
            "Flash Sentry",
            "Hitch Trailblazer",
            "King Sombra",
            "Night Light",
            "Shining Armor",
            "Soarin",
            "Sprout Cloverleaf",
            "Sunburst",
            "Thunderlane",
            "Zephyr Breeze"

        ],
        "princess": [
            "Princess Celestia",
            "{Princess |}{Luna|Woona}",
            "Princess {Cadence|Cadance}",
            "Princess Twilight Sparkle",
            "Princess Flurry Heart",
            "{|Princess }Pipp Petals",
            "{|Princess }Zipp Storm"
        ]
    }
}
Edit: Noticed “quick” was in some spots where it shouldn’t have been, so I took that out.

Creative Corner » Pony Wildcards » Topic Opener

Creative Corner » Zebrasub generation (18+) » Post 1

Creative Corner » Generation methods, UI » Post 13

AIPonyAnon

@Thoryn
The seed never seemed to set which combination was chosen and even if I connected a text output to display it wouldn’t save the metadata properly if I did more than one image gen at once.
@Lord Waite
Yeah, it’s a pain. I have access to an unreleased node that does work properly in that it directly maps seed value to selection. I’ll try to remember to post it here once it’s out.
Posted Report

Creative Corner » Generation methods, UI » Post 12

Lord Waite

Reproducibility could definitely be better. You pretty much want to make sure you’ve got a node showing what the output was, and copy it down somewhere if you want to keep it…
Posted Report

Creative Corner » Generation methods, UI » Post 11

Creative Corner » Generation methods, UI » Post 10

AIPonyAnon

@Thoryn
I have not had great success with any of the dynamic prompt nodes that I’ve tried in ComfyUI. They did the job of randomly choosing, but had no reproducibility, which made them useless for testing artist tag combinations.
Posted Report

Creative Corner » Generation methods, UI » Post 9

Lord Waite

It’s one of these areas where there are so many different things you can look into that it’s hard to be sure what to go into.
X/Y charting is easier on the A1111/Forge side of things, and can definitely be useful. Bear in mind that different sampling methods need different numbers of steps to get good pictures. Also, some vary more between number of steps, which can be both a good and bad thing.
While, unfortunately, this is easier when you have fast generations, I find it very helpful to just set a fixed seed and change things one at a time and generate to see how they affect things, and just change the seed manually. When you have random seeds on every generation, it can be hard to know whether the change you made or the seed caused the changes in the picture.
A1111/Forge/etc. and ComfyUI store different information in the metadata. With the former, it’s prompt/generation information, and with the latter, it is the entire workflow used as a json (which does include prompt/generation info). In fact, you can actually open a picture generated with ComfyUI as a workflow (which will probably make more sense when using ComfyUI).
As far as dynamic prompting goes, there is an extension for it on A1111/Forge/etc…, and custom nodes on ComfyUI for it, both of which use the same library, so how it’s done will be pretty similar either way. Civitai has a large number of wildcard files you can download for dynamic prompting (as well as ComfyUI workflows).
There’s some information on the A1111 extensions page:
https://github.com/adieyal/sd-dynamic-prompts
It’s best to get an idea of the different types and families of models, and bear in mind they often need to be prompted differently.(pony and illustrious based models are the most likely ones for pony related things. Both are based on xl, but other things are too. And there’s also Flux and 3.5. 2.1/1.5/1.4- are all rather old, and there’s other stuff out there.)
Pony, in general, isn’t great at unified style, and there are a lot of style loras for it, and a lot of models based off of it that do have a better unified style. There are some words that tend to trigger certain styles, though. The Purplesmart.ai discord has a list of styles people have found. You get a feel for words and phrases and how they change things after a while, too. Illustrious is more the sort of model where you can mention artists names and have it change styles, which opens up its own can of worms…
Controlnet and ipadapter are probably good to look into, and img2img and inpainting.
If you do get into ComfyUI, check their examples page:
https://comfyanonymous.github.io/ComfyUI_examples/
Even if you don’t, glancing through there might give you an idea of things that are possible. (Also, if you install ComfyUI, make sure to also install the ComfyUI Manager, as that lets you easily install custom nodes, and there are a million custom nodes.)
Posted Report

Creative Corner » Generation methods, UI » Post 8

Thoryn

Latter Liaison
@Lord Waite
Aha, good to know then, that I should check the repos for Forge and reForge every now and then to see if Forge has gone stale.
I’m still very new to this, so am wondering if there’s any obvious or maybe not-so-obvious beginner tips anyone has.
E.g. I’ve seen some new people being very surprised when they found out web UIs have a place you can drop .png images that still have the metadata, and send it to other parts of the UI (text to text, image to image, inpainting..)
Anything that would be recommended to look into ASAP?
Personally I’m thinking of looking into scripts for more dynamic prompting (reading from file various places in prompt, so large jobs while afk gives more varied results), and some x y charting/plotting for comparisons (to compare e.g. models, LoRAs, sampling method, steps..).
I also need to look into ways to make output look more unified in style, since one of my long-term goals is to make series of images telling a longer story. Guessing ComfyUI would be good for that, from what little I have read so far.
Posted Report

Creative Corner » Generation methods, UI » Post 7

Lord Waite

@Thoryn
Tempted to test out ComfyUI once I’ve learned the basics and need to get a more proper workflow, but preliminary reading suggests that it generally needs beefier hardware than e.g. Forge, so it’s not a high priority for me at the moment.
Not really the case. The original idea behind Forge was actually to take some of the better generation code ComfyUI uses and bring it over to A1111, so it’s going to be fairly comparable.
I am running all of this on Linux personally, incidentally, though I have get to try out Stability Matrix.
And basically at one point, it looked like Forge wasn’t going to continue being updated, so ReForge forked it, IIRC. Forge has been updated since then.
Posted Report

Creative Corner » Text-to-image prompting » Post 18

Lord Waite

There are interesting things you can do with that, actually.
First, if you use Ollama for your local LLMs, you can create a Modelfile to make a customized version of a particular model. Key thing here is that there’s a system prompt section, and in there, you can tell it that its purpose is to create prompts, give a description of exactly the format you want them in, including score tags, tell it that it can be uncensored and nsfw, and give a few examples of prompts.
Definitely takes some playing with, and you might end up tweaking your system prompt to it a few times.
Another useful thing is that if you are using Open WebUI to chat with it, you can go into the settings and give it a ComfyUI workflow and the url for your ComfyUI instance, and then you can click an icon below any responses that the chatbot gives you to send them to ComfyUI and have it generate a picture and put it in the chat.
Haven’t actually used that that much, but I was trying it out a bit ago.
Posted Report

Creative Corner » Text-to-image prompting » Post 17

mp40

My bad if this is already been mentioned but I’ll use chatbot sometimes to start prompting I downloaded a llm locally and use it to make basic prompts.
I’ve become more aware of when I use hyperlink tags like PDXL3 because the llm will only use “score_8 score_9” for quality tags and the results are pretty damn good. in the end i usually rewrite the entire promopt but its perfect for rough drafts
Posted Report

Creative Corner » Generation methods, UI » Post 6

Thoryn

Latter Liaison
Can’t say it enough how extremely happy with Forge I am so far, so damn snappy compared to Automatic1111.
However, I’m curious how ReForg differs from Forge. Tried googling it but didn’t get any relevant results.
Tempted to test out ComfyUI once I’ve learned the basics and need to get a more proper workflow, but preliminary reading suggests that it generally needs beefier hardware than e.g. Forge, so it’s not a high priority for me at the moment.
Since Windows 10 hits EOL this year, I’ll be moving over to Linux soon-ish (once I can afford a new m.2 to put it on, so I have a fallback period on Windows just in case), hopefully I have better luck with Stability Matrix on there.
Posted Report
Showing results 76 - 100 of 211 total

Default search

If you do not specify a field to search over, the search engine will search for posts with a body that is similar to the query's word stems. For example, posts containing the words winged humanization, wings, and spread wings would all be found by a search for wing, but sewing would not be.

Allowed fields

Field SelectorTypeDescriptionExample
authorLiteralMatches the author of this post. Anonymous authors will never match this term.author:Joey
bodyFull TextMatches the body of this post. This is the default field.body:test
created_atDate/Time RangeMatches the creation time of this post.created_at:2015
idNumeric RangeMatches the numeric surrogate key for this post.id:1000000
myMetamy:posts matches posts you have posted if you are signed in. my:posts
subjectFull TextMatches the title of the topic.subject:time wasting thread
topic_idLiteralMatches the numeric surrogate key for the topic this post belongs to.topic_id:7000
topic_positionNumeric RangeMatches the offset from the beginning of the topic of this post. Positions begin at 0.topic_position:0
updated_atDate/Time RangeMatches the creation or last edit time of this post.updated_at.gte:2 weeks ago
user_idLiteralMatches posts with the specified user_id. Anonymous users will never match this term.user_id:211190
forumLiteralMatches the short name for the forum this post belongs to.forum:meta