Use with Loras/Hypernetworks/Textual Inversions
You will find that Divinity can get some incredible outputs when paired with the above. I highly, highly recommend trying this model out with something like sadxzero's SXZ Bloom Hypernetwork to start:
https://civitai.com/models/4845/sxz-bloom
I also 100% recommend trying out LuisaP's Loras which are absolutely incredible as well! https://civitai.com/user/luisa_pinguin
The What & Why
This model's size is a big one coming in at 7.17gbs. (I know, I'm sorry, lol)
Divinity does best with booru tags though you can absolutely experiment and mix up typical prompt tags to see what you get out of this model. I usually use this model for semi-realistic images and close-ups/portraits. This model is absolutely wonderful if you want to go for landscapes and scenery with that concept art vibe to it.
The model does pure anime and it does it very well as it will be the default output style you will get if you don't get specific with your prompting. NSFW was not in my mind at all though but if you want that kind of thing feel free to try and show off your images.
Prompt Stuff?
"((best quality)), ((masterpiece)), ((realistic)),"
This basic starter prompt is going to be your absolute BEST FRIEND when using Divinity and the generations you get will hopefully satisfy you. Pair that alongside good prompting skills and anything else you can take advantage of will do you wonders in my honest opinion.
VAE?
There is a VAE already baked into the model so there is no need to worry about needing one, I promise.
Highres Fix Will Serve You Well
I absolutely advise using the highres fix on AUTOMATIC1111's repo to also aid in maximizing the potential quality of your images. You can also play around with SD Upscale as well after you're finished, your choice. Every image shown here uses the highres fix at 1.5x base resolution.
Negative Prompt?
"(low quality, worst quality:1.4), (watermark), (bad_prompt_version2:0.8)"
You don't really need to do a 200 token to 300 token negative prompt unless for some reason you want to, to each their own. I recommend trying out Nerfgun3's Textual Inversion which helps weed out some bad generations, particularly on the hands.
https://huggingface.co/datasets/Nerfgun3/bad_prompt
The Wrap Up
Hopefully you guys and gals out there will enjoy Divinity, if you need any tips or suggestions or just want to chat feel free to contact me on Discord.
Immobilis Dignitas#8258
Divinity is a highly specialized Image generation AI Model of type Safetensors / Checkpoint AI Model created by AI community user immobilis_dignitas. Derived from the powerful Stable Diffusion (SD 1.5) model, Divinity has undergone an extensive fine-tuning process, leveraging the power of a dataset consisting of images generated by other AI models or user-contributed data. This fine-tuning process ensures that Divinity is capable of generating images that are highly relevant to the specific use-cases it was designed for, such as anime, landscapes, concept art.
With a rating of 5 and over 7 ratings, Divinity is a popular choice among users for generating high-quality images from text prompts.
Yes! You can download the latest version of Divinity from here.
To use Divinity, download the model checkpoint file and set up an UI for running Stable Diffusion models (for example, AUTOMATIC1111). Then, provide the model with a detailed text prompt to generate an image. Experiment with different prompts and settings to achieve the desired results. If this sounds a bit complicated, check out our initial guide to Stable Diffusion – it might be of help. And if you really want to dive deep into AI image generation and understand how set up AUTOMATIC1111 to use Safetensors / Checkpoint AI Models like Divinity, check out our crash course in AI image generation.
1.0 Standard