Resized to 73% of original (view original)
| Prompt | masterpiece, best quality, very aesthetic, very awa BREAK jz235, gritty, faux traditional media, painting \(medium\) BREAK 1girl, nagatoro hayase, tanlines, petite, smug, nipples, small breasts, pussy, completely nude, on side, after sex, cum in pussy, cumdrip, excessive cum, cum pool <lora:noob10vpred-butterchalk-unet only-october6v2-step00002464:1> |
|---|---|
| Negative prompt | sound effects, lowres, (bad), text, error, fewer, extra, missing, worst quality, jpeg artifacts, low quality, watermark, unfinished, displeasing, early, signature, extra digits, artistic error, username, scan, bruise on face, bruise |
| Sampler | Euler |
| Seed | 4285304574 |
| Steps | 20 |
| Cfg Scale | 5 |
| Model Hash | ea349eeae8 |
| Advanced Sampling Enabled | True |
| Advanced Sampling Mode | Discrete |
| Controlnet 0 | Module: tile_resample, Model: xinsir controlnet tile [4d6257d3], Weight: 0.35, Resize Mode: ResizeMode.INNER_FIT, Processor Res: 0.5, Threshold A: 0.5, Threshold B: 0.5, Guidance Start: 0, Guidance End: 0.45, Pixel Perfect: False, Control Mode: ControlMode.BALANCED, Hr Option: HiResFixOption.BOTH, Lora hashes: noob10vpred-butterchalk-unet only-october6v2-step00002464: 0x74b6821b5c |
| Denoising Strength | 0.8 |
| Discrete Sampling | v_prediction |
| Discrete Zsnr | True |
| Model | noobaiXLNAIXL_vPred10Version |
| Schedule Type | Normal |
| Size | 1152x1638 |
| Version | f1.2.0dev-v1.10.1RC-previous-955-g456f2771 |
- ‹ prev Search: status:any next ›
Anonymous 10/08/25(Wed)16:58:48 No.8740277
>>8740263
>Could you take a quick look to check what I've done wrong?
>Model: 291H_v10
also try supplementing the prompt with additional styling tags like
>gritty, faux traditional media, painting \(medium\)
also note that you're not going to actually get filmgrain from image gen. It just doesn't work, mostly due to VAE nonsense. If you want filmgrain you'll need to do postprocess in photoshop or krita or whatever(this should literally take 2 seconds to do once you know how to do it).
also your dataset being half horse porn probably isn't doing you much of any favors. especially the one with four panels, or the other image with a bunch of text scrawled all over it(even if they're properly tagged that stuff will bleed or just not train properly).
that said I wasn't entirely 100% happy with the first run of butterchalk I did on my end but I'm also not too interested in focusing too hard on it
https://files.catbox.moe/e2e72n.png