I really don’t like this new update to the model.
Just being honest — the old version was way better for what I was doing.
I’ve been using this model mainly to create anime-style characters for my stories. Since the update, everything feels off. The art style is totally different — more exaggerated, stylized in a weird way, and it doesn’t match the classic anime look I was working with before. The characters I built and grew attached to just don’t look right anymore.
Yeah, I get that even with the old model there were small differences from image to image — like a character might be a little shorter, taller, or their expression would shift slightly. But that was fine. It actually gave them more personality and made them feel unique without overdoing it. It was subtle. It wasn’t all over the place like it is now.
With the old model, I could generate the characters how I wanted — and I’d end up with over 200 stills of the same character that I could use for storytelling. It was almost like building a storyboard. And if one didn’t come out quite right, I could just tweak the prompt a little and get a better version without losing the look.
Now with this update? I can’t even get the same character to look consistent from one image to the next. Each picture looks exaggeratedly different, like it’s a different art style or an entirely different character. The faces, proportions, vibe — all over the place. They go way off model.
And worst of all, I’ve lost that fine control we had. The brackets like ((( ))) actually worked before — to fine-tune things like bust size, height, body shape, emotions, you name it. Now, even without the brackets, it feels like the model is jumping to extremes for no reason. I don’t want my characters to be cartoonishly altered just because I said “tall” or “serious.”
All I’m saying is: for the love of god, bring the old model back. Or at the very least, give us the option to use the old one. Some people might love this new version, and that’s fine. But for people like me who were relying on the older model for consistent, story-based character generation, this is a real setback.
Anyone else feeling the same?
Sort of but not really.
I see that, after 2 days of checking it out there are some negatives. I haven’t tried to make identical characters like u yet, tho i notice it has few main styles so people often have a certain look and don’t differ much. The old version recognizing 800+ different artist styles was incomparably more variation than the like 3 main styles now. That’s the biggest negative I personally see yet.
Positives tho: I could post an image somewhere people share images and not have it ridiculed for being super low resolution with jpeg artifacts. So new version is successfully ‘modern’. Also I was able to make a complex comic with all sorts of specifics which wouldn’t have been possible b4. I think it will take some time to switch how we mentally approach creating an image. Less gently guiding character creation with vague words and more specifics and detail and complex concepts. Another plus is for people generating nudity who no longer have weird model inabilities to do certain anatomies.
I would say mostly: we right now have tons of img generators built along the lines of how SD1.5 generated images that are now using the same structure to do Flux. I’m making a modern gen to work with flux, I know at least one other person who is, and I think more ppl should be. Give it a month and some of a new era of image gens should be coming online. Imo this rough patch after the update but before gens able to handle it are out is to be expected. That’s why even the biggest negative I see (the lack of diversity in styles) I’m not really paying attention to; have to get generators built on this update out first before we can really say what this update can or can’t do.
I get where you’re coming from — and yeah, for people looking for high-res or doing experimental stuff, I can see why the update feels exciting. But for users like me, who need consistent character generation and traditional anime styles for ongoing projects, this change has been a major blow.
I’ve already built whole characters and stories using the older model. I wasn’t trying to push boundaries — I was just trying to make my characters look the same across different scenes and poses. Now, they don’t even resemble themselves. Every image looks exaggerated or off-style, and no amount of prompt tweaking fixes it. And while I understand that maybe more tools will eventually come out that work well with this model… that doesn’t help right now.
The loss of variety and bracket control is also a massive downgrade. Before, I could fine-tune things subtly — different expressions, slightly different heights or builds — and still have it feel like the same person. Now it’s like I’m rolling a dice every time I hit “generate.”
So yeah, maybe it’s “modern,” but it’s not practical for storytelling at the moment. I wish we could just have the option to use the old model alongside the new one. Let the people who like this new style enjoy it, but don’t leave the rest of us behind.