I’ve tested it out, works pretty well. However, I would like to have an ability to quickly add my saved tags to images during image upload and tag editing processes.
Posts
Search Results
Site and Policy » Site performance » Topic Opener
Over about the past week or so, the Tantabus.ai main page, and image pages, have been extremely slow to load, occasionally taking well over a minute to complete. I’m on a 200-Mbps cable connection, and I’ve performed a variety of ping and speed tests to confirm the problem lies not on my end. Other ’boorus are performing as expected*.
In addition to the Webpages loading slowly, image file downloads also sometimes slow to the low-KiB/s range, and result in taking several minutes to complete, or (more lately) complete failure.
Note that this is NOT ALWAYS the case. It’s a very intermittent problem; one image page might load fine, the next could take two or three minutes, and it seems to have absolutely NO relation to the file size or content type presented.
This ‘booru is the only one I visit that has been having these problems, to an extent that interferes with using the site. I’m not mad or anything; I just wanted to bring it to attention, so that it can be fixed for all. :)
* Derpibooru is having its own (Cloudflare-related?) hiccups now. No time-outs, just slow page-loading.
–> And just now, I had an SSL handshake failure (error code 525). Hmm….
FYI, my system is running Arch Linux, using Firefox 146.0.1 – yes, I know, I need to update to 147. :P
EDIT: I’ve been made aware by support on Discord that there’s a buttload of maintenance going on with Cloudflare right now…but mainly in Europe? I’m in the Midwest, though. Weird.
Creative Corner » Tips on making pictures more interesting » Topic Opener
Do you want to create something more interesting than a standard picture of a single character in a generic pose? Here are some tips that can help you spice things up:
Style
Current models support various tags that can vastly change the image’s style, like
traditional art, pencil sketch, colored pencil drawing, watercolor painting, acrylic painting, pixel art, 2d, flat colors, black outlines, modern art, lineless, vector, limited palette, monochrome, greyscale, black and white, anime style, etc. Try out different tags and see how they affect the picture. You can even increase or decrease the tag weight to get different results - for example, (watercolor painting:1.4) is much stronger than (watercolor painting:0.6).CFG also affects the style - the lower the value, the more artistic, desaturated and messy the result is. Test which values work best for you. For Illustrious/Noob based models, it would be around 1.5 - 3.5. For Pony-based models, it would be around 3.5 - 7.5.
Some examples of different styles:
Narrative
Implanting a narrative can give some actual impact to a picture that would be pretty forgettable otherwise. When creating something, ask yourself these questions: What kind of an idea am I trying to get across? What am I telling the viewer? Are things shown in the image clearly helping with this?
When deciding on a narrative, you can draw inspiration from one of the MLP episodes, a fanfic you’ve read, a meme you’ve seen, or something you have experienced yourself. The possibilities are endless. You can also use text or dialogue, but you would probably need to insert it yourself, since most models can’t generate text reliably.
Some examples of narratives inspired by different things:
Meme:
MLP fanfic:
MLP song:
MLP episode:
Personal musings:
Magazine cover:
Characters
A trend I’ve noticed in a lot of generations - a character with a generic smile and a basic pose, just standing around and doing nothing. No interesting pose or expression, no unusual outfit, no interaction with their surroundings or other characters, just nothing that would capture the viewer’s attention (apart from the character’s anatomy, perhaps).
Think of these questions: Why would your character be in this place? What would they do there? Does it fit their personality? What kind of expression would they have? What kind of outfit would they wear (if any)? Does the whole composition make sense? All of these have to fit in the narrative you have chosen.
There is a lot you can achieve with a single character, but having two or more gets even more interesting. Are they friends, enemies, strangers, lovers? Maybe there is a group dynamic you would like to portray? Unfortunately, generating more than one character can be tricky, especially if they are interacting with each other. Techniques like img2img, inpainting and regional prompting combined with some manual editing can get you very far, but it requires work and patience.
Some examples:
Holiday theme, matching outfits, character interaction, different expressions:
Unusual outfit and a goofy expression, tied by a short story:
Friendship/rivalry:
Cozy Glow being Cozy Glow:
Two ponies stargazing while enjoying each other’s company:
Creative Corner » Liquid Rainbow Effect » Topic Opener
Liquid Rainbow
Foreword
So, around a week ago I started playing around with Illustrious. And coming from PonyV6, I fast learned that it can be a lot more “creative”. Yes, a PonyV6 models knows ponies better, but Illustrious, in my opinion, knows more “concepts”. As I do with new toys, I played around and found the effect mentioned in the title:
Resources
We need 3 things:
- An Illustrious model. I worked with this one. But I tested it with this one, too. Both work, for others you have to try yourself.
- The “Velvet’s Mythic Fantasy Styles” Lora in version 2 for illustrious. (This Lora is for a detailed anime style. Any Lora adding detail could work, I like this one.)
- The “LeIsT0 | Shiiro’s Styles” Lora. (This one helps making the colors “pop”.)
The Prompt
The prompt for the Twilight picture looked something like this:
QUALITY WORDS, 1pony, my little pony, pony, feral pony, unicorn, Twilight Sparkle, casting magic at viewer, shiny fur, limited palette, (glow, liquid glowing magic, glowing liquid purple magic)++, intense expression, dramatic lighting, strong depth of field, wind-blown hair, gritty realism, cinematic shot, dutch angle, (simple background, black background)++, dust particles, dynamic pose, dynamic composition, foreshortening, blurry edges, MORE QUALITY WORDSSome may be confused about the “++” signs. I work with Invoke, which takes “-” and “+” signs as weights. A “++” is around 1.2. So if you use a different UI, you will have to try around a little. The important parts are
simple background, black background for the mat black background to play with. If I lowered the weights, the loras added stuff into it.The effect itself is achieved with
limited palette, dramatic lighting, glow, liquid glowing magic, glowing liquid purple magic. Limited palette makes the AI use as few colors as possible, what is then overridden by the “purple magic” part with a higher weight. Muting any other color beside the one mentioned. If you replace “purple” with “rainbow”, you get the colors of the rainbow. Easy.Now the lora weights. I used both with good results at 0.8. The trigger words are not in the prompt since Invoke has its own way to add loras that doesn’t need them. If you add a lora, it gets activated, that’s it. You find the trigger words on the civitai page.
I used Euler A at 25 steps with a CFG around 7 for all the pictures here.
That should be it, really. Have fun!
Important
The more in detail you describe the character in the prompt, the more detail gets added, and with that: color. If you want good “shade play” be as brief as possible. (That’s why it works better with known characters since you only have to add their name most of the time.)
More Examples
A few more raw examples:
Side Note
If you use the Fantasy Lora at 0.8-1.0 and the “LeIsT0” Lora at 1, with ‘limited palette’ and no rainbow liquid you get this comic looking style:
If you use the Fantasy Lora at 0.8-1.0 and the “LeIsT0” Lora at 1, with ‘limited palette’ and no rainbow liquid you get this comic looking style:
Creative Corner » Samplers » Post 2
Background Pony #B443
Ooh, nice!
Unless the model specifies it needs a specific sampler, I tend to opt for Restart since it’s like a mini hi-res fix.
Do you have a similar resource for Scheduling type? In most of my tests the differences are minimal, as art timing. Except of some Illustrious based models where not using the recommended adds a lot of “noise patterns” (waves, flowers, swirls…) in empty areas.
Site and Policy » Opening Thread » Post 176
Woo, thanks for running my script; I am glad to see the comments on the tantabus now.
Also, are the comments still being imported? Because I noticed a few missing comments, which I never had issues with during my tests. If someone could send me the
processing.log file, that would be great so I can fix that potential issue.Here is a small example that does not have the comments imported:
Tantabus & Derpi
And to sanity check that comment does exist on my test environment:
Tantabus & Derpi
And to sanity check that comment does exist on my test environment:
I’ve also noticed that I neglected to check for deleted comments to prevent them from being imported, so any deleted/banned comments are now visible again as raw text. I can adjust the script to fix that issue.
Creative Corner » Generation methods, UI » Post 10
@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.
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.
Creative Corner » Generation methods, UI » Post 7
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.
Creative Corner » Generation methods, UI » Post 6
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.
Creative Corner » Post Your AI Art! » Post 37
I am personally interested in the best one can get from the AI pipeline from prompt to output image without human input of any kind after the initial prompt, so nothing I post uses any post-generation editing, not even inpainting.
Sounds like the approach I’m trying to take. I’ll likely have to learn some digital painting eventually, for minor touch-ups, or conformity for dialogue boxes and fonts etc, but for the most part, I want to do it with prompting.
Reading your description of how you do it, and your elaboration of it on your user page, makes me hopeful that you’ll share any knowledge you can in an appropriate thread in this forum.
Remember seeing extensions for Automatic1111 that could do wildcard prompting, but I never got around to test it. I assume it’s possible with others as well.
Creative Corner » Text-to-image prompting » Post 11
@Lord Waite
For the record, I’m using RTX 3070 TI with 8GB of video memory.
For the record, I’m using RTX 3070 TI with 8GB of video memory.
Good point on the fact that LoRAs also use some memory.. best to avoid if possible.
Don’t have the sampler named Uni_pc on my setup, so I kept it on DPM++ 2S a paired with Karras - as when I did a test with all the samplers (same prompt, seed etc), that was one of the fastest (Euler A was slightly faster and what I’ve been using when experimenting with Automatic a couple times before, but lately I have seen more instances of DPM++ 2S a in the wild, so figured I’d give it a try).
I copied your prompt exactly, and it actually gave a decent composition this time!
At 1024x1024 and 15 steps, it took 13 minutes… maybe this is passable for getting the composition, then I can use img2img to flesh things out? I should start experimenting with the pipeline soon, like only loading one or two LoRAs at a time (especially for kinks and concepts I know the model can’t do at well or at all), figure out grid-editing/prompting, inpainting etc.
At 1024x1024 and 15 steps, it took 13 minutes… maybe this is passable for getting the composition, then I can use img2img to flesh things out? I should start experimenting with the pipeline soon, like only loading one or two LoRAs at a time (especially for kinks and concepts I know the model can’t do at well or at all), figure out grid-editing/prompting, inpainting etc.
score_9,score_8_up,score_7_up,score_6_up,score_5_up,score_4_up,
source_pony,rating_safe,
Flurry Heart lying under Christmas tree,ceiling,window,Christmas decorations on walls,presents,pony,filly,cute,star on top of tree,
source_pony,rating_safe,
Flurry Heart lying under Christmas tree,ceiling,window,Christmas decorations on walls,presents,pony,filly,cute,star on top of tree,
Steps: 15, Sampler: DPM++ 2S a, Schedule type: Karras, CFG scale: 6, Seed: 40318147, Size: 1024x1024, Model hash: 67ab2fd8ec, Model: ponyDiffusionV6XL_v6StartWithThisOne, VAE hash: 95f26a5ab0, VAE: sdxl_vae.safetensors, Version: v1.10.1
Creative Corner » Text-to-image prompting » Post 10
@Thoryn
If you have a weak GPU, then how about trying out Forge WebUI? It looks almost exactly like A1111, but it should be much faster, especially on a weak GPU. I don’t know about newer versions, but I’m using a release from 02.2024 and it’s working great. I have GeForce 3060Ti with 8GB VRAM and I can generate four 1024x1024 pictures in slightly above one minute. Pretty much no OOM, errors or crashes.
https://github.com/lllyasviel/stable-diffusion-webui-forge/releases
If you have a weak GPU, then how about trying out Forge WebUI? It looks almost exactly like A1111, but it should be much faster, especially on a weak GPU. I don’t know about newer versions, but I’m using a release from 02.2024 and it’s working great. I have GeForce 3060Ti with 8GB VRAM and I can generate four 1024x1024 pictures in slightly above one minute. Pretty much no OOM, errors or crashes.
https://github.com/lllyasviel/stable-diffusion-webui-forge/releases
There are also other UIs like ReForge or ComfyUI, I’d recommend testing different options to see what suits you best. Stability Matrix is great for installing and maintaining multiple UIs.
https://github.com/LykosAI/StabilityMatrix
https://github.com/LykosAI/StabilityMatrix
Creative Corner » Text-to-image prompting » Post 8
@Lord Waite
Thanks for the tips. Have used 512x512 because I have basically the bare minimum of capable hardware (with that res, it With 75 prompt chunks, it usually takes me almost 5 for ~23-25 steps, and 10 minutes for 32-35.. and you’re saying I need to quadruple the res, oof..
(I promised myself not to throw more money at expensive GPUs as I am broke and have stopped gaming, but a 5090 starts to look more appealing the more I mess around with AI).
Thanks for the tips. Have used 512x512 because I have basically the bare minimum of capable hardware (with that res, it With 75 prompt chunks, it usually takes me almost 5 for ~23-25 steps, and 10 minutes for 32-35.. and you’re saying I need to quadruple the res, oof..
(I promised myself not to throw more money at expensive GPUs as I am broke and have stopped gaming, but a 5090 starts to look more appealing the more I mess around with AI).
You’re correct that I’m using Automatic1111 by the way.
(Have pondered alternatives, as the cmd window spews errors left and right even on a fresh and up-to-date version, but it’s the devil I know right now..)
(Have pondered alternatives, as the cmd window spews errors left and right even on a fresh and up-to-date version, but it’s the devil I know right now..)
I will avoid LoRAs, clean out the negative prompts (and add things only as needed), up the res to 1024x1024 and do some testing.
Thanks again for all the input, really appreciate it.
Thanks again for all the input, really appreciate it.
Creative Corner » Model and Lora (and related) Discussion » Post 19
I tried out NoobAI-XL v-pred 1.0 yesterday, https://huggingface.co/Laxhar/noobai-XL-Vpred-1.0:
Wildcarded through a fair few combinations of artist styles when generating this image. Judging from that, it seems to have a better understanding of style and texture than the eps-pred version. It also seems to be slightly worse at fine detail, but that may just be a product of its better understanding of style. It needs further testing with artist tags that do highly detailed works to see if that is a limitation of the model.
Wildcarded through a fair few combinations of artist styles when generating this image. Judging from that, it seems to have a better understanding of style and texture than the eps-pred version. It also seems to be slightly worse at fine detail, but that may just be a product of its better understanding of style. It needs further testing with artist tags that do highly detailed works to see if that is a limitation of the model.
Site and Policy » Opening Thread » Post 122
Does the dump include tag metadata like categories, descriptions, implications, and aliases?
Yes, it does. It also includes the tag changes.
I’ve already exported the difference between derpibooru and tantabus, which can be found here. However, this isn’t an import tool, just a list.
If the admins/mods want, I could also have that option in the import tool. I have the script ready already because I imported all the tags to my dev environment for testing, which has all the categories, descriptions, implications, aliases, and a few others. I just need to modify it to search for existing tags and update them.
Site and Policy » Opening Thread » Post 114
I’ve created the import tool and successfully tested it in my development environment. Here’s a preview of a small sample of the data it migrated:
Features:
- Comment Importation: Seamlessly imports comments from a CSV file exported from the Derpibooru nightly dump into a PostgreSQL database.
- Comment Updates: Updates existing comments in the database if they already exist, utilizing a mapped JSON file created during the first run.
- User Anonymity Handling: Manages user anonymity for users who do not exist in the system.
- Comment Suffix Details: Adds suffix details to comments based on user existence status.
- Configurable Import Delay: Processes comments with a configurable delay between each import.
- Batch Importing: Imports comments in batches, optimizing performance.
- Callback Features: Includes callback functionality for various columns such as:
- Match users based on usernames (since IDs differ across sites)
- link comments to images based on descriptions containing the source URL and ID.
Additionally, if preferred the import user can be anonymous:
Just need to do some edge casing & checks and I’ll share the code, and it’ll need to be run by someone who has PostgreSQL access.
Site and Policy » Opening Thread » Post 102
@AIPonyAnon
That should be added now, thanks!
That should be added now, thanks!
@BigBuggyBastage
As mentioned, that’s OK; ideally there’d be something that lets people know about it to avoid error or confusion.
As mentioned, that’s OK; ideally there’d be something that lets people know about it to avoid error or confusion.
@IWTCIPP
Tagging (and filtering by it) would be good, but I’m not sure there’s the bandwidth for that; as in, uploaders probably wouldn’t tag their own stuff with “negative tags”, so there’d need to be a fair number of active users going around curating each image. It would be an additional workload, albeit distributed on more people that just mods, and avoid “unnecessary deletions”, but could also fail if there aren’t enough people going at it.
Tagging (and filtering by it) would be good, but I’m not sure there’s the bandwidth for that; as in, uploaders probably wouldn’t tag their own stuff with “negative tags”, so there’d need to be a fair number of active users going around curating each image. It would be an additional workload, albeit distributed on more people that just mods, and avoid “unnecessary deletions”, but could also fail if there aren’t enough people going at it.
It’s something worth revisiting down the line, though.
@Background Pony #BAD9
It’s my hope that this site will have a fair amount of activity and that people can come here when they want to get a wall of nice pics, well organized pics, without dealing with the terror that is CivitAI or DA’s search/filtering systems.
It’s my hope that this site will have a fair amount of activity and that people can come here when they want to get a wall of nice pics, well organized pics, without dealing with the terror that is CivitAI or DA’s search/filtering systems.
For comments, there is a possibilty (but not a guarantee) that we’ll be able to copy those over.
For funding, everything is separate from both derpibooru and furbooru (and those from eachother). It makes accounting for what’s using (and costing) what, but also gives some surety that if one site gets hit by an orbital laser, the others will be fine. Currently the cost of running this place is fairly low, definitely not a concern, so I’m happy to carry it for the foreseeable future. If costs increase (which would probably be a symptom of a Good Thing), then I’ll look into the same sort of stuff used on derpi (donations and non-network ads).
@FlatterDeN_AI
@Zerowinger
Unfortunately the subject of AI (particularly for art) has a lot of… “passion” in it. I can understand where a lot of traditional artists are coming from, I’ve been in several AI focused discords (and other places) where people boast and relish in artists losing commission money, going unemployed, being made obsolete, etc. While those people may be a minority, their hateful or cruel comments are what leaks out and gets heard by a lot of artists, that then assume everyone that’s ever touched an AI generator is personally out to leave them destitute in the streets.
@Zerowinger
Unfortunately the subject of AI (particularly for art) has a lot of… “passion” in it. I can understand where a lot of traditional artists are coming from, I’ve been in several AI focused discords (and other places) where people boast and relish in artists losing commission money, going unemployed, being made obsolete, etc. While those people may be a minority, their hateful or cruel comments are what leaks out and gets heard by a lot of artists, that then assume everyone that’s ever touched an AI generator is personally out to leave them destitute in the streets.
I really don’t know what to do about that; I wish people could just calm down and that assholes would go to therapy or deal with their anger some other way, but I’m just some guy.
We’d really hoped to have launched far far sooner and had more tech and policy issues sorted and tested out before now, but several major Real Life events got in the way that took priority. I’m hoping we can work out the kinks quickly and smoothly enough, but do ask for some patience while things are still in progress. If you have concerns or suggestions, we do want to hear them; there’s stuff we’ve already changed from it, and don’t see a reason to just set everything in stone now.
Creative Corner » Post Your AI Art! » Post 27
This was a test image of Flurry Heart I did to see if Imagen knew what she looked like, it produced a really nice output I think. Doesn’t show her wings, though it could be argued they’re under the dress.
Site and Policy » Opening Thread » Post 89
@Background Pony #89EC
One of our registration related services was down for a while, but should be back and working again (tested now to be sure). It’s possible Privacy Badger isn’t liking reCaptcha, but that’d be odd as it’s a fairly popular thing… That or the browser is out of date and reCaptcha fails it, I’m not sure.
One of our registration related services was down for a while, but should be back and working again (tested now to be sure). It’s possible Privacy Badger isn’t liking reCaptcha, but that’d be odd as it’s a fairly popular thing… That or the browser is out of date and reCaptcha fails it, I’m not sure.
@Background Pony #7E45
We’ll import images slowly over the next few days to not flood out other uploads.
We’ll import images slowly over the next few days to not flood out other uploads.
Creative Corner » Model and Lora (and related) Discussion » Post 11
@tyto4tme4l
Technically, there’s a new version available, but it’s based on v-pred NoobAI, which is still in training. If I understood correctly, the author plans to remake it once NoobAI finishes training.
Technically, there’s a new version available, but it’s based on v-pred NoobAI, which is still in training. If I understood correctly, the author plans to remake it once NoobAI finishes training.
I only did some cursory testing on it and, ngl, it felt undercooked, whereas Zoinks just clicked with me instantly.
Tagging Discussion » General "Whoa this tag is missing stuff!" Thread » Post 9
@Background Pony #6406
The site is open source. Knock yourself out and let us know when you have it tested. In the meantime, we’ll do it the way that works, with thanks to the people doing the heavy lifting.
The site is open source. Knock yourself out and let us know when you have it tested. In the meantime, we’ll do it the way that works, with thanks to the people doing the heavy lifting.
Creative Corner » Post Your AI Art! » Post 11
@Zerowinger
It is, and I knew I had to upload it when I generated it. I’ve always been fond of the two of them, and testing generating them and Twist is one of my typical tests for a new pony model.
It is, and I knew I had to upload it when I generated it. I’ve always been fond of the two of them, and testing generating them and Twist is one of my typical tests for a new pony model.
(I also test generating named stallions because ponies like Prince Blueblood, Fancypants, Trenderhoof, and such really should be in there, but generally don’t have much luck on that one…)
Creative Corner » Post Your AI Art! » Post 2
Oh I love that one of chrysalis! I recall seeing it and I actually went and tested making a similar concept with Luna and then Celestia to match
Luna: https://tantabus.ai/images/25280?q=prompter%3Amontaraz13
Celestia: https://tantabus.ai/images/25281?q=prompter%3Amontaraz13
Luna: https://tantabus.ai/images/25280?q=prompter%3Amontaraz13
Celestia: https://tantabus.ai/images/25281?q=prompter%3Amontaraz13
This is probably the one I dedicated the most time to (Light bondage/nsfw warning): https://tantabus.ai/images/6784?q=prompter%3Amontaraz13
Showing results 1 - 23 of 23 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 Selector | Type | Description | Example |
|---|---|---|---|
author | Literal | Matches the author of this post. Anonymous authors will never match this term. | author:Joey |
body | Full Text | Matches the body of this post. This is the default field. | body:test |
created_at | Date/Time Range | Matches the creation time of this post. | created_at:2015 |
id | Numeric Range | Matches the numeric surrogate key for this post. | id:1000000 |
my | Meta | my:posts matches posts you have posted if you are signed in. | my:posts |
subject | Full Text | Matches the title of the topic. | subject:time wasting thread |
topic_id | Literal | Matches the numeric surrogate key for the topic this post belongs to. | topic_id:7000 |
topic_position | Numeric Range | Matches the offset from the beginning of the topic of this post. Positions begin at 0. | topic_position:0 |
updated_at | Date/Time Range | Matches the creation or last edit time of this post. | updated_at.gte:2 weeks ago |
user_id | Literal | Matches posts with the specified user_id. Anonymous users will never match this term. | user_id:211190 |
forum | Literal | Matches the short name for the forum this post belongs to. | forum:meta |