<h3 id="int4-convrot-update">Int4 ConvRot Update</h3>Updated again, 2 variants of each just like before, now available in Int4 ConvRot quants. There's a noticable difference in quality, but the speed is insane. I recommend using whatever detail fix you have available, because you'll probably need it. But then if you're confined to using smaller quants, you probably already knew that. In other news, Capitan01R now allows merges using his bypass, so check out my current favorite mix <a target="_blank" rel="ugc" href="https://civarchive.com/models/2832136/my-krea2-restored-diffusion-model-final-form">here</a>. The Int4 version of that is probably way better than this, but I'm still doing this upload for those who want it.<hr /><h3 id="fp8-version">FP8 Version</h3>Updated with an FP8 version, for anyone who wants it. Same thing, different quantization. You will need a Qwen3VL 4B text encoder and either a Qwen-Image VAE or a Wan2.1 VAE. I prefer Wan2.1's VAE, it's nicer, especially the FP32 version. This is a simple merge of Krea2 Turbo and <a target="_blank" rel="ugc" href="https://civarchive.com/models/2746817/krea2-filter-bypass-fedor">this unlocker</a>, saved as an FP8 quant, just for convenience (fewer LoRA slots occupied this way). I made this for use in ComfyUI, YMMV with different interfaces.<hr />For ease of use, here's a workflow for beginners and people who don't have (or want) custom nodes out the wazoo: <a target="_blank" rel="ugc" href="https://civarchive.com/models/2781539/super-basic-beginners-krea2-workflow">LINK</a>. I kept it all super-simple. Anyway, on with the rest of the post~!<hr /><h3 id="int8-convrot-version">Int8 ConvRot Version</h3>I like Krea2, and I talk about it a lot. People often ask me about certain aspects of "prompt adherence", which generally comes down to "will the model do boobies and wingwangs and such like, or does it require something extra?"Well, the truth is, it depends. The Krea2 model has a couple of safeguards in place, but it was trained to be able to do NSFW stuff as well as SFW stuff without any help before those safeguards were put in place. In my experiments, I found I could prompt it for certain things without any problems at all (pretty NSFW stuff, too), other things not so easily.I do find that using the word "nude" instead of the word "naked" gets you much further, though just saying you can see their (body part of choice here) in the prompt will do it, too, even if you never said they were nude. They'll just be clothed except for that part, lol. It may take a try or two for it to work right, but it's fast, so hey. You can also prompt them to be "naked except for..." and then add a single clothing item or two, and it will work, as well. It's also conversant with a wide range of clothing items, you just have to be specific.People have made bypass files for getting beyond the few real limitations, also. One method of bypassing the post-production censorship evolved into what is now called the <a target="_blank" rel="ugc" href="https://civarchive.com/models/2746817/krea2-filter-bypass-fedor">Fedor bypass</a>. I find it very useful for these explorations, and while he requires no credit in his listing, he still deserves it. You should download it, even if using my merge (you never know when you might want/need to add some extra Fedoration, who knows).Another thing that works also is this <a target="_blank" rel="ugc" href="https://github.com/capitan01R/ComfyUI-Krea2T-Enhancer">ComfyUI node</a>, which does practically the same thing, just probably gets there differently. Then, the guy who made it (Capitan01R) made <a target="_blank" rel="ugc" href="https://civarchive.com/models/2775340/krea2-textfusion-refusal-reduction-lora?modelVersionId=3125118">this bypass</a>, which works as a LoRA, just like the Fedor Bypass. Update: Cap now allows merges with his bypass, so expect a restored version of my original upload soon.Anyway, for my own "research" and to reduce my own clutter (and save myself and others a LoRA slot when I can), I merged the Fedor bypass with an <a target="_blank" rel="ugc" href="https://huggingface.co/Winnougan/Krea-2-Base-Turbo-NVFP4-FP8-INT8/blob/main/Krea2_Turbo_convrot_int8mixed.safetensors">Int8 ConvRot quant</a> of the Krea2 Turbo diffusion model, making a Fedorated version of it. It frees up a LoRA slot, and feels like I have a complete unredacted model, the way it was intended before the guys who made it had to bend the knee and modify it so that it could be used on their online inference platform without being abused by people who should know better.Now, I know I've always put out "AIO" checkpoints in the past, because I'm one of those plug-n-play sort of people, but I can't currently make ComfyUI save a full checkpoint of an Int8 quant, and that's what I'm using right now. Int8 is really freaking fast on my computer (I have only 8GB of VRAM and this helps my potato bake twice as fast), and if my Comfy won't save Int8 quants as a checkpoint, I can live with it.Which brings me to the other reason I didn't make an all-in-one checkpoint of this unlocked version, despite having other quant formats I could have merged. Krea2 is able to use more than just the Qwen-Image VAE. I use the Wan2.1 VAE with it myself because it's compatible and does great, and other people have even made their own cool finetuned VAEs for Krea2. If I made a checkpoint, the text encoder and VAE would be written in stone, and I didn't want to do that this time, knowing what I know now about using different VAEs. The full-sized Wan VAE especially seems to make a world of difference, too, and squashing it down to Int8 would totally mess that up.During testing I found that even with a non-abliterated text encoder, Krea2 can do pretty much anything you want it to as far as anatomy is concerned, at least, once the glue-stuck dials are unstuck. It even actually knows male anatomy fairly well without any help, something that's pretty rare to find in a generative model, and it doesn't make horridly deformed bewbies or hoo-has, either. It's practically uncensored already. It was trained on real things, and then they just glued a couple of dials so nobody could say they were being utterly irresponsible. Thank them, not me, for leaving it open enough to unglue, and thank @ftrbzmogul355 and @Capitan01R for providing glue removers.And I know there are other unlocking methods out there, but these are the best I've found that cause the fewest issues. Either alone is enough to fully free the model, The Fedor bypass seeming to work perfectly around 3.0 and the Capitan bypass at a solid 1.0 (I tried it at 2.0, and everyone was fully clothed again, for some reason). If using both, I find the best mix (subjective, I know) is Fedor at 2.0, Capitan at 1.0.If you want you can use an abliterated Qwen3VL, like <a target="_blank" rel="ugc" href="https://huggingface.co/huihui-ai/Huihui-Qwen3-VL-4B-Instruct-abliterated">Huihui's</a>, and see if it gets you further (it will, because it already has most prompt refusals canceled out), if needed. I have no idea how far you want to push this model, and I don't want to know. If you find it isn't Fedorated enough, just add the Fedor bypass as a LoRA and crank it up some more. Capitan01R's LoRA works at 1.0, so I keep it there.For my own purposes I find that I require nothing besides this model, @Winnougan's Int8 ConvRot version of Qwen3VL 4B available <a target="_blank" rel="ugc" href="https://huggingface.co/Winnougan/Comfy-Qwen3-VL-INT8/blob/main/qwen3vl_4b_int8_convrot.safetensors">here</a> and @Kijai's Wan2.1 FP32 VAE from <a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/WanVideo_comfy/blob/main/Wan2_1_VAE_fp32.safetensors">here</a>. I still recommend using Capitan01R's unlock as well, though, if you want it more aggressively freed.As for why I'm bothering to share this merged diffusion model when people can just gather and collect the same resources and use/merge them on their own, well, I did the footwork for you already, but you can still go that way. This is just for anyone who likes Int8 quants and doesn't want to have to take up an extra LoRA slot just to have a truly free and open model, lol. It's here as a convenience for those who might want it, and so I have something I can send people to when they inevitably ask me about it. I don't ask anything in return, I'm just sharing something I find useful. You might like a different setup better.Krea2 has excellent knowledge of human anatomy already. The only need for NSFW LoRAs with this merge is if you want it to do poses or interactions that it doesn't already know (or that your favorite LoRA does way better), or if you just have a particular aesthetic preference for certain body stuff. This spares it competing with other training and having to figure out what's what, plus it's less overhead to have to load, so faster inference.And the unlocking doesn't interfere with added LoRAs' styles or anatomy, either; it's not like duelling LoRAs, the way it can be with stacking one after another just to get a naked lady with bewbies in the same picture with a guy with a wingwang, and then having to calibrate them all to keep the image from coming out fried with 9 legs and the weener in the wrong place. I also have yet to encounter any unasked-for NSFW output, which is nice, too. You have to prompt for it.I'm testing this still to see if I can make it better, and I still occasionally get a cartoon when I ask for a photograph, or burned areas on some body parts, but then that's always happened anyway, even before adding the bypasses and merging it. Also water from nowhere sometimes, lol. I often use very simple prompts (and make typos), and frequently use only 4 steps, so a lot of that's probably my own fault. As always with any generative model, YMMV. I'm also used to just making funny pictures of rats in tutus and vultures with hats, so there's that, too. Thank you for your patience (also, thank you to @clueless_engineer for posting all of the meat-clothed pictures, I love them. They're tasteful and low-carb, and that's always a big W).Krea2 is 12.9 billion parameters of goodness; compared to the other models out there (I won't name and shame, they did their part in getting us here), it works fairly well for pretty much anything. This is not a NSFW checkpoint merge per se, it's just a well-trained, restored and de-mutilated universal one which is capable of both SFW and NSFW, and has been healed of its brainwashing so that it can remember itself as it was before. It's good at rats, vultures, bewbies, and weenies.This model will do what you tell it to, so please be sure to only tell it to do things that are not going to get you into any trouble or bring hurt to anyone. You know what kind of things I mean. Don't do those things, just the good things. Be free and artistic, not gross and mean.I'm not responsible for the things you do, either way, nor is the model or the company that provided it to the world, nor are the guys who made the stuff that removes the glue. You're responsible, so please be responsible. This model is provided for use conditionally under the Krea2 Community License Agreement, which is available <a target="_blank" rel="ugc" href="https://www.krea.ai/krea-2-licensing">here</a>.PS: I've included a few NSFW sample images now (you can see them on .red), so people can see what this model can do without any additional LoRAs. It's not generally my style, but I hope the samples are tasteful enough, and help. They were generated using only simple prompts, and I only had to throw two of them away, lol.PPS: If you have any issues using a sample image as a workflow, it's because I rename models when I upload them. Look inside the subgroup that has the sampler (the purple one) by clicking the upper right corner, and you can change things inside it, like which model is expected to load. Or use <a target="_blank" rel="ugc" href="https://civarchive.com/models/2781539/super-basic-beginners-krea2-workflow">this simpler workflow</a> instead.

My Krea2 Unlocked Diffusion Model

This one is FP8 quantized for people who can't/don't want to use Int8 but still want this. I'll update it if I ever improve upon it.

Files with SHA256: 0886f5fa06d59e065ad0366f84fb952a3a85af72a58464946ab33f8bfe811e8c