Name: CineVisionXL by SoCalGuitarist
Rating: 4.99 (284 reviews)
Author: socalguitarist

thick thighs

sexual situations

physical violence

disturbing

extremist

hanging

hate symbols

nazi party

breasts out

strapless leotard

revealing clothes

weapon violence

female swimwear or underwear

male swimwear or underwear

partial nudity

self injury

white supremacy

adult toys

graphic male nudity

illustrated explicit nudity

nudity

sexual activity

graphic violence or gore

graphic female nudity

vore

bestiality

corpses

breast out

one breast out

huge breasts

gigantic breasts

huge butt

wide hips

covered nipples

hair over breasts

no panties

sitting on face

convenient censoring

anal

dildo riding

downblouse

peeing

oral

oral invitation

emaciated bodies

hentai

porn

bukkake

cumshot

cumdrip

cum in pussy

cum on face

cum on hair

cum on body

cum on tongue

cum on hands

cum in mouth

cum on pussy

eat_cum

fellatio

implied fellatio

after fellatio

triple fellatio

autofellatio

pov fellatio

fucked silly

presenting ass

presenting pussy

spreading legs

spread legs

swimwear

lingerie

bikini

latex clothing

TRPCError

Like the work I do and want to help support me? <a target="_blank" rel="ugc" href="https://www.buymeacoffee.com/socalguitarist">Buy me a coffee to say thanks</a> or <a rel="ugc" href="patreon.com/user?u=33487230">Support me on Patreon for exclusive early access to my models and more! </a><a target="_blank" rel="ugc" href="https://discord.gg/SKrfwWNa9u">Join us on SCG-Playground where we have fun contests, discuss model and prompt creation, AI news and share our art to our hearts content in THE FLOOD! 💖💖💖</a><hr />This is CineVision XL, a new trained model I have built with several of the awesome folks from my Discord channel SCG-Playground. This model has been trained on widescreen (16:9, 21:10) formatted images and can produce stunning "straight out of a movie" scenes that are coherent and beautiful.CineVision XL is capable of both SFW and (mild) NSFW output. As with all of my other models, tools and embeddings, CineVision XL is easy to use, preferring simple prompts and letting the model do the heavy lifting, but also can understand complex scene building to a surprising level of detail (try it!)Special thanks and shout-out to @Blink, @Klinter, @humblemikey for their direct contributions to this model, and big thanks and love ❤️❤️ to my ultra-patient model testing crew on Discord who I'm constantly barraging with new LoRAs and internal builds to try to break. 🤙🏽🤙🏽<a target="_blank" rel="ugc" href="https://www.instagram.com/socalguitarist_/">Follow me on IG!</a><a target="_blank" rel="ugc" href="https://www.threads.net/@socalguitarist_">Follow me on Threads!</a><a target="_blank" rel="ugc" href="https://discord.gg/SKrfwWNa9u">Chat with me on Discord</a><a rel="ugc" href="patreon.com/user?u=33487230">Support me on Patreon</a>

CineVision 2.3.0 Public Release NotesHello folks, this one has been in the barrel for about a month and a half now, with multiple test revisions and multiple layers of training. All told, CineVision received about 120k total training over close to 9k training images. CineVision is a movie-focused model that will default to cinematic realism styles as it's normal output. It is very responsive to many different types of cinematic styles, terminology and scene setup. It is very coherent and responsive to your prompt, and works best with natural language prompting like in my example images. CineVision, despite its name, is actually quite dynamic as a model - It can create photographs, cartoons, movies, weird scribbles, coloring books - you name it, it will probably work. With SD3 drop coming right around the corner, this will probably be my last big SDXL update for this model (and my other models, which have "final XL versions" coming soon). I hope you enjoy it, it's my daily driver and remains so even as other new models come out. I'll be dropping a lightning version as well, and it also works great with the new Hyper loras too (several of the samples I've added it, though I think Hyper is a bit too hot on CFG)Hands and medium distance objects are quite good now, especially if you use SAG or PAG (or both). Changelog 4/25/24 <ul><li>5 trainings for about 120k total steps on several thousand training images.</li><li>3000 cinema stills handpicked and upscaled with ultrasharp to &gt; 1024 and &lt; 2048 on it's longest side</li><li>2000 photos from a photography dataset</li><li>2000 images from LAiON-POP dataset</li><li>1600 MJ aesthetic art dataset (my own library, hand selected)</li><li>350 anatomy training images</li><li>All captions done with GPTV/LLaVA-1.6</li></ul>Known Issues<ul><li>Men's genitalia has come a long way (no pun intended) but still struggles with generating convincing penises. Best to use a lora and inpainting if you're really wanting accurate male physiology.</li><li>smudgy effects with some movie styles - this can show up sometimes due to training on 24FPS content early on. It's mostly been quashed now, but can still show up if you're really pushing the old fashioned analog movie look (think 50's style technicolor)</li></ul>Note - All of my sample images for this version are using <a rel="ugc" href="https://github.com/pamparamm/sd-perturbed-attention.git">Perturbed Attention Guidance (PAG)</a>, FreeU, self attention guidance, latent modifier, DynamicThresholding. All but PAG are built into <a rel="ugc" href="https://github.com/lllyasviel/stable-diffusion-webui-forge">ForgeUI</a>, which I heartily recommend over A1111.

CineVisionXL_2.3.0

Version 1.5.0 - Initial Public Release NotesI'm so excited to be releasing this model out to the world! Several hundred hours worth of work went into building this model out, and while it's not perfect, I think it's at least as high quality as my other model lines DynaVision, ProtoVision, and NightVision.CineVisionXL was Initially tuned with several hundred synthetic movie themed images that were generated using MidJourney, DALL-E and NightVisionXL and "themed" using Lightroom for cinematic feeling and styling. Next, a low value merge (at 10%) with PixelWave in coordination with HumbleMikey as well as the LoRAs Anthroid from Klinter and CaricatureXL from Blink (Thanks guys!). On top of this, another 3 rounds of training using cinematic "establishing shots" style synthetic scenes, and working in of my own internally trained detailer and cinematic LoRAs to get us where we are now.Known Issues<ul><li>hands</li><li>eyes and teeth need more work, sometimes teeth on lips or eyeshine weirdness can happen</li><li>male genitalia is pretty much non-existent (use a LoRA)</li></ul>

Release-v1.5.0-BakedVAE

CineVisionXL by SoCalGuitarist

Type	Checkpoint Trained
Stats	3,657
Reviews	Overwhelmingly Positive (528)
Published	Nov 4, 2023
Base Model	SDXL 1.0
Hash	AutoV2 AB4FDDB198