AI generated zako

[Zanduras]

Recently tried to generating some locally and its amazing what you can do now. Haven't done so since the original Stable Diffusion that struggled with any scene more complex than a common pose. Inpainting and controlnet steps was too finicky for my casual use.

The Krea 2 model is really good for complex scenes by having nuance prompting with natural text with some story hints rather than sledgehammer global booru style tag prompting. Specifically using fine-tuned model Ultra, it recognizes the complex scene and style prompts better than the base model in general.

Here's a couple random assortment I made using comfyui and a messy reference workflow I've currently settled on. Would like to be able to one shot generate a delmo without getting into LoRA or fine tuning, but so far I can only reinforce the outfit details with a reference image with if it got trained on it.


View attachment 179132View attachment 179133View attachment 179135View attachment 179136View attachment 179137View attachment 179139

Update:
Interesting, scaling up the resolution lowers fixed a lot of distortions so no real need for a 2nd upscale pass to maybe correct issues. At 1440p, I'm able to get mostly clean results that were impossible to get right on 720p.
View attachment 179141 View attachment 179142
prompt for reference (same flow without the outfit embedding):


Update 2:
Cleaned up the workflow and added to this post. Removed the useless composition prompt block that I'm going to try again later. Since I'm not getting a precise outfit and the outfit block bleeds style (anime outfit causes anime output), I added a reinterpretation pre-processing block to reinterpret the original reference. With the style more aligned, the uniform signal (multiplier block) can be set higher to better inject the outfit in, not perfect though.

View attachment 179191

View attachment 179208 View attachment 179209

View attachment 179192View attachment 179193

Update 4:
Found out the CLIP model is where prompt cross-referencing ("attention") occurs, when I tried to use append clip vectors it just had this free floating idea of the prompt, which did happen to work when all characters wore the same uniform but stopped working when I had a unique character.

Was able to insert both a character and zako uniform prompt but definitely tricky and a hack. Maybe I'm missing some magic sauce system prompt, but Qwen VL and Krea 2 was really struggle to reconcile my inputs (probably my fault).

Here's an example (not the best) and workflow. It did do much better in some generations but had anatomy issues.

Golden delmo takes care of their government agent problem directly:
View attachment 179210 View attachment 179211 View attachment 179212

Tried out Ultra and having some fun with generations. Been a while since I've been so engrossed. AI is very very far from perfect but wow, it's starting to become only limited by ones own degenerate imagination.

Not amazing for stories yet but for single gens yah it's good enough.

 

[SexyAsh]

Watch it here:AshZako

[R-18G]逃脱｜预告

小蕾从实验台上苏醒。击败眼前的修女医生后，她的逃脱才刚刚开始——地狱火组织的修女特工、修女禁卫军接连出现，通往出口的路上，还有更多阻碍等待着她。 先放几张截图，和大家分享这次的制作进度！ 目前已完成约 6分钟，最后约1分钟的剧情正在制作中。成片预计约7分钟，计划9月底发布。 这次

www.pixiv.net

Xiaolei wakes up on an examination table. After defeating the Sister Medic, her escape is only beginning. Hellfire’s Sister Agents and Sister Praetorian Guards stand in her way—and reaching the exit will take more than she expected.
Here are a few screenshots to share how production is coming along!
Around six minutes are complete, with the final minute of the story still in production. The finished video should run approximately seven minutes, with release planned for late September.
Expect the distinctive uniforms, female combat, and continuing storyline you know from my work. I’m using AI tools in production, while carefully refining the story structure, shot selection, character consistency, action continuity, and editing, sequence by sequence. I want every shot to feel like part of the story.
I’m putting together a substantial episode, with plenty of attention to the details. Thank you for following and supporting my work. See you in late September!
Which scene are you most looking forward to from these screenshots?
小蕾从实验台上苏醒。打败了眼前的修女医生后，她的逃脱才刚刚开始——地狱火组织的修女特工、修女禁卫军接连出现，通往出口的路上，还有更多的阻碍等待着她。
先放几张截图，和大家分享一下这次的制作细节！
目前已完成约6分钟，最后约1分钟的剧情正在制作中。预计成片约7分钟，计划9月底发布。
这次依然有大家熟悉的时装设计、女性对抗和连续剧情。制作了使用了AI工具，而故事编排、镜头选择希望的角色一致性、动作一致性与时装节奏，都在逐段调整、反复修改，让每个镜头都属于卡拉故事。
这次希望能让大家看得过瘾，也看得到细节上的用心。感谢一直关注和支持的朋友，9月底见！
这几张截图里，你最期待哪一幕？