Like the work I do and want to say thanks? Buy me a coffee or Support me on Patreon for exclusive early access to my models and more!
NightVisionXL started as a lightly trained, photography focused SDXL model that specialized in photographic output. Over the past few months, NightVision has received dozens of trainings and tunings chasing the dragon that is pleasing aesthetic output. With the release of version 8.X, I feel like I'm much closer to my goal of effortless beautiful output that is aesthetically pleasing without requiring confusing negatives or comma separated nonsense.
NightVisionXL speaks natural language. Prompt it like you would describe a scene to a human being, and you'll be surprised by how well it can understand and comprehend your prompts.
I'm trying something new with V8 of NightVision, and I hope the results show in the output. I continue to scrape beautiful imagery that I think will train well, and this particular update is no exception with hundreds of beautiful license free AI images that I've sourced from MJ, Dalle3, Civit, Playground and my own outputs. On top of that, I'm also utilizing HF datasets for license free photography and the LAION-POP dataset, filtered to remove low quality images prior to training. Captioning is done using my own Spicy Burrito captioning tool (available to my Patreon subscribers on discord!) that creates incredibly verbose and effective captions that are essential for good coherent output.
On a final note, I've had multiple requests to "turbo-ize" this model. I have no plans to mix turbo into NightVision, as the licensing on NightVision is nice and open right now with the SDXL licensing, and I'd like to keep it that way. NightVision isn't fast, but I promise you the longer gen times are worth the wait!
This is NightVision XL, a lightly trained base SDXL model that is then further refined with community LORAs to get it to where it is now. NightVision XL has been refined and biased to produce touched-up photorealistic portrait output that is ready-stylized for Social media posting! NightVision XL has nice coherency and is avoiding some of the weird body issues and biases that are starting to plague some of the other photorealistic models. Further, NightVision XL produces rich deep blacks and great evening/night time scenes. It can also produce ridiculously bright output as well!
NightVision XL is capable of both SFW and NSFW output. As with all of my other models, tools and embeddings, NightVision XL is easy to use, preferring simple prompts and letting the model do the heavy lifting for scene building.
NOTE - This version includes a baked VAE, no need to download or use the "suggested" external VAE.
WARNING - DO NOT USE SDXL REFINER WITH NIGHTVISION XL
The SDXL refiner is incompatible and you will have reduced quality output if you try to use the base model refiner with NightVision XL.
Follow me here by clicking the heart ā¤ļø and liking the model š, and you will be notified of any future versions I release. I also need your help with feedback, please please please post your images and your honest feedback below, I will use your feedback and your output to help guide future revisions!
Patreon coming soon!
Description
Nightvision XL Alpha 0.6.5.0 Release Notes 08/16/23
This new release should hopefully fix (or mostly fix) the nipples on the outside of clothes issue that was showing up in 0.5.2. It should also have more facial variety, better response to cinematic terms like LUT and color temperature, and should maintain the nice creamy dark images that NightVision is designed to produce. Finally, and this is a big one, I've added more facial variety to the output, which should stave off some of the biases that are starting to show up more prominently in other realistic models (mostly Asian / female / body shape bias). I'm also continuing to try to improve eyes and teeth.
0.6.50 Changelog
Added new LORAs
pfac (Realistic_Human_Face_portraits_XL)
TeethXL
Rebalanced nurlens_SDXL_V0.1, MovieStillXLV7 Black_Sa_May and SDXL_TeethXL
Known Issues
Hands
Penises
teeth and eyes at a distance can look off
NSFW may be a bit harder to make happen, and women's anatomy may have taken a small hit with trying to fix the nipple issues. Can still produce artful nudes just fine, but porn output may be affected.
FAQ
Comments (16)
Nightvision XL Alpha 0.6.5.0 Release 2 08/16/23
This new release should hopefully fix (or mostly fix) the nipples on the outside of clothes issue that was showing up in 0.5.2. It should also have more facial variety, better response to cinematic terms like LUT and color temperature, and should maintain the nice creamy dark images that NightVision is designed to produce. Finally, and this is a big one, I've added more facial variety to the output, which should stave off some of the biases that are starting to show up more prominently in other realistic models (mostly Asian / female / body shape bias). I'm also continuing to try to improve eyes and teeth.
0.6.50 Changelog
Added new LORAs
pfac (Realistic_Human_Face_portraits_XL)
TeethXL
Rebalanced nurlens_SDXL_V0.1, MovieStillXLV7 Black_Sa_May and finenude_v0_2a
Known Issues
Hands
Penises
teeth and eyes at a distance can look off
NSFW may be a bit harder to make happen, and women's anatomy may have taken a small hit with trying to fix the nipple issues. Can still produce artful nudes just fine, but porn output may be affected.
Fantastic Model and your right about porn output. Impossible for me to achieve anything close.
Excellent model. I think in some respects it already surpasses any 1.5 model, such as the detail and consistency of facial features, perspectives and even the damn hands, and in other respects it's just a matter of refining the model further.
Stable diffusion XL certainly has a lot more potential than version 1.5, seeing what you model trainers are capable of when this new version has only recently been released.
Thanks for sharing your work.
Hey, are you using high res fix or upscaling on your images and are you using adetailer?
I use HRF for the demos, no face detailers or any other extra fixes or processes. It's all in the metadata šš¼
I'm having a problem rendering with this model only, it gives me this error:
RuntimeError: The size of tensor a (2048) must match the size of tensor b (768) at non-singleton dimension 1. is there something I can do to fix this? thank you!
Are you rendering at 1024x1024?
@tomasreichmannĀ no, tried different dimensions and prompts, will give it another try today to see what I get.
My guess is that you included a Lora not trained for SDXL in your prompt. Remove the Lora and the error should dissapear
@AURORA_HĀ sdxl must used by verison>1.5,you should git pull
@JannisDZĀ I have also had this happen when trying to use certain controlnet models. My most current info is that controlnet does not (yet) work at all with XL.
@JannisDZĀ I have had this happen to me before but no cuz I tried the same model with no loras and same outcome
@xue_renĀ can you please explain more?
@AURORA_HĀ He's saying make sure your UI is up to date. Get the latest updates to A111 or whatever, and try again if you still have problems. ComfyUI is also a good alternative, but takes a few to learn. Much more resource flexible though.
@AURORA_HĀ All you need to do is make sure that you do not use any SD1.5 resources with SDXL model. They are not compatible. Must download SDXL specific controlnet/lora etc.
@AURORA_HĀ you can see your UI version at bottom of the browser ,keep your UI version is equal or greater than 1.5 and when you use sdxl donāt use anything based on other model training.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.