By gerogero
Updated: October 16, 2026
New update for content ratings: every image you generate on BetterWaifu's own models now gets a content rating: General, Sensitive, Questionable or Explicit.
Social posts get one too, based on the images in them. You can use ratings to sort your history, and to blur anything spicier than you want on screen, whether it is your own generation or someone else's post.
Here is what each rating means.
RatingWhat it coversTypical examplesGeneralCompletely safe for work. Nothing suggestive.Fully clothed characters, portraits, landscapes, everyday scenes.SensitiveMildly suggestive. Not something you want on a work screen, but no nudity.Underwear, bikinis, tight bodysuits, revealing outfits, cleavage or panty shots.QuestionableModerately NSFW. Clearly sexual, but stops short of explicit sex.Partial nudity, heavy sexual subtext, ecchi scenes.Explicit18+ only. Full-on adult content.Sex acts, full graphic nudity, hardcore erotic images.UnratedNot checked yet.Older images, videos, some models, and posts containing any of those (more on that below).
When an image finishes on one of our anime models, it goes through an open-source anime image tagger before it lands in your history. The tagger looks at the picture itself, not your prompt, and gives a score for each of the four ratings. Your image gets filed under whichever one scores highest. If two are tied, it goes to the spicier one, so a borderline image never gets labelled milder than it is.
The four levels are the same ones Danbooru uses, written there as g, s, q and e. The tagger learned from art rated on that scale, so if you already know booru ratings, ours mean the same thing. Sensitive is the one people get wrong most: it is not "a bit spicy", it is the step between safe and NSFW. A bikini pin-up is Sensitive. Once there is nudity or a clearly sexual scene, it is Questionable or higher.
A few things worth knowing:
It is automatic, so it is not perfect. The tagger is very good at the obvious cases and less sure about borderline ones. A swimsuit shot can land on Sensitive or Questionable depending on pose and framing.
It rates the image, not the prompt. If you asked for something explicit and got something tame, the rating follows what is actually in the picture.
It never changes what you can generate. Ratings are labels for filtering. They do not block, hide or delete anything on their own.
Ratings only exist for images that went through the tagger. That leaves a few groups without one:
Older history. Tagging started in late August 2026. Images from before that have no rating.
Videos. Video clips are not rated yet.
Some models. Images from BWAI Edit and SeeDream are not run through the tagger.
Unrated does not mean safe. It just means we have not looked yet. That is why the browsing level below treats Unrated images and posts as above every level when you turn it on.
A post is rated by its images, and the most explicit one wins. A post with four General images and one Explicit image is Explicit, because that one image is what you would see when you scroll through it.
If any image in a post is Unrated, the whole post is Unrated. That mostly happens with posts made from older images or videos. When you post an image that was never tagged, we tag it right after you post, and the post's rating updates on its own a moment later.
On the History page, open Filters and use the Rating section. Tick one or more of General, Sensitive, Questionable, Explicit or Unrated to show only those. Nothing ticked shows everything.
If you browse somewhere other people can see your screen, open the Content filters panel and use the Browsing level slider. Drag the eye to the highest rating you want to see. Anything above it is blurred until you tap it:
Your generations on History and Create.
Social posts on the feeds, tag pages, profiles and the post page itself.
Choosing Explicit turns the blur off. Tapping a blurred post shows it everywhere it appears for the rest of your visit, so you do not have to unblur the same post twice.
This one is rolling out gradually, so you may not see it yet. It is saved on your device, so set it again on your phone if you want it there too.
These are two different things, and they sit in the same panel. The category filters (furry, futanari, gay and so on) hide whole kinds of posts from your feeds. The browsing level is about how explicit something is: it blurs your generations and posts above the level you pick, whatever their subject. Use either, both or neither.
Your generations are private in your history unless you post them. Once you post, the post gets a rating like any other, so people who set a lower browsing level will see it blurred until they tap it. Nobody sees a rating label on your post, only the blur.
No. A blurred post still shows up in feeds and still counts views and likes. It only looks blurred to people who chose a lower browsing level, and it is one tap away.
No. Ratings are labels. They do not cost anything and they do not block generations.
Not yet. The tagger decides on its own. If you see a lot of clearly wrong ratings, tell us in Discord with a few examples.
Historical images are still being rated. As older images get a rating, the posts that contain them update too.
If you have a browsing level set and the image came from a model that is not tagged (BWAI Edit, SeeDream, or any video), it is Unrated, and Unrated counts as above every level. Tap it to show it.
Want to try it? Generate a few images in the image generator, then open your History and filter by rating, or set a browsing level and scroll the feed. For better prompts, the Danbooru tag groups guide is a good place to start.
Get new guides and NSFW AI updates in your inbox. No spam, unsubscribe anytime.

Complicated desired outputs = Complex prompts with mix of natural language and tags [postcard=89ntmto] Complex prompt...

This guide was created to bring inspiration to this visual vocabulary. There is a short description for each pose so ...


This guide is intended to get you generating quality NSFW images as quickly as possible with Automatic1111 Stable Dif...