13 Best Stable Diffusion Models – Easily Exceeding Midjourney


If you click on a link and make a purchase, I may receive a small commission. As an Amazon affiliate partner, I may earn from qualifying purchases.
Read our disclosure.



In the development of the Henmix_Real model, the creators have taken great care to make it appealing and accessible to both Western and Asian users. A key focus has been placed on enriching the generation of dark expressions, infusing them with vibrancy and dynamism. As a result, the overall tone of the model’s output tends to be darker, a deliberate choice that complements and shines particularly well on dark backgrounds.

From users:

“thanks for this great model, best model so far with my test for both man and woman, and great for multiple people too, deserve more attention!!”

“This is by far my favorite model for people, but I would really love if it had a bit more diversity. It is very asian biased, and does not do other skin tones and shapes quite as well. It’s definitely harder to do fair skin, even using pale prompting and negative asian.”

Check out: Best photography and realistic AI art models

Realistic Vision


Realistic Vision is available through Civitai and is in continuous development. Use prompts like RAW photo, (high detailed skin:1.2), 8k uhd, dslr, soft lighting, and high quality to get the best results. For the negative prompt, use words like deformed iris, deformed pupils, semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, and anime:1.4.

Usually, in photo-realistic images, persons/characters’ pupils tend to get distorted, or the subject might look a bit too cartoony, use negative prompts to counter that.

From users:

“wow I thought chillout was the best model and then I found this”

“This is my favorite ckpt model. It generates the most beautiful and realistic women Ive found so far of 7 other models Ive downloaded.”

Deliberate V2


Deliberate V2 is available through Civitai and has been downloaded more than 160,000 times. The model needs very accurate prompts to give you the best possible results.

From users:

“Da best model ever! I tested a lot of models, but I always came back to this one. Thank you!”

“Best model on Civit.ai!”



The MajicMIX AI art model leans toward Asian aesthetics, showcasing its unique artistic flair. Constantly evolving, the diffusion model employed in this AI is regarded as one of the top-notch Stable Diffusion models available. With its exceptional capabilities, this model excels in generating realistic-looking images that possess a distinct cinematic touch, adding a touch of artistic elegance to each creation.

From users:

“Thx for nice work, this is my most favorite model.”

“this is a great model but doesnt seem to work with any TI, anyone having the same issue?”



DreamShaper is free to use on Hugging Face, but if you want to download it for yourself, you can find it from Civitai. It also includes VAE baked into it. DreamShaper is continuously developed and the latest version can create some of the best AI art.

From users:

“My favorite model updated! V5 is here! WOW! Thanks!!”

“One of the best mix, as all the previous versions.”

NeverEnding Dream


NeverEnding Dream is available on multiple AI art generator sites such as Getimg.ai, Sinkin.ai, and Mage.Space. This model complements DreamShaper but does not have Dreamlike in the mix. Best for cosplay images, anime pictures, good-looking people, realistic animals, etc.

From users:

“First of all, this model is amazing. Creates amazing images with good quality. The only problem I’m having is getting a different face. The checkpoint tends to recreate a similar face every time (like a young/childlike face). Tried with “mature face” prompt and negative “child, young) and still creates that similar face. Still working on the propmts trying to get a more mature face on the models.”



DucHaitenAIart Stable Diffusion model is perfect for cartoony and anime-like character creation. You can get it from Hugging Face. DucHaiten is updating the model and creating new ones, so be sure to follow him on Patreon.

Always add to the prompt: masterpiece, best quality, 1girl or 1boy, realistic, anime or cartoon, 3D, pixar, highly detail eyes, perfect eyes, both eyes are the same (optional), smooth, perfect face, hd, 2k, 4k , 8k, 16k



One of the best and diverse high-quality anime AI art models for Stable Diffusion. You can get Counterfeit-V3.0 from Civitai. Some users have trouble using the model though as there are no clear instructions on how to use it.

From users:

“3.0 is very good model. But I have only 1 matter of generated images. Girl’s face is very detailed and beautiful but,their nose are too instability. They are Interrupted and increased in number.”



Probably one of my favorites Stable Diffusion models currently available at Civitai. GhostMix diffusion model generates some of the best-looking anime art, and the art style feels like a mix of reality and fantasy combined with anime aesthetics.

The creator Ghost_Shell is constantly improving the model for it to make even better AI art.

From users:

“Hi Ghost_Shell, amazing model, I just discovered it. I agree with you it is the best out there right now for all this 2.5D style work.”

“For people who are having trouble getting this to work: whatever version of Stable Diffusion you’re using needs to support Loras. I was trying and trying, and couldn’t get anything close to the example images. Then I updated my Easy Diffusion with the ability to use Loras, and now it works perfectly. Beautiful model, thanks for your hard work!”

Dark Sushi 2.5D


If you are into anime art, Dark Sushi Mix and the updated version Dark Sushi 2.5D are excellent Stable Diffusion models to have.

From users:

“Absolute masterpiece model !”

“I’ve got a problem with hands. No matter what I do there’s always a problem with hands. Having more than 3 hands or mutated hands etc. Do you have a suggestion for this problem? Great model btw”



UnstableInkDream checkpoint model is already at its 7.5 version and is in constant development. 7.5 version has been split into three different models as it was impossible to combine all versions together. There’s 7.5balance, 7.5-anime, and 7.5-photo available for download.

RunDiffusion FX Photorealistic


RunDiffusion has released a checkpoint-trained RunDiffusion FX Photorealistic AI art model that is able to produce some of the best-looking AI photography art. With a perfect blend of simplicity and functionality, this cutting-edge model produces a wide array of captivating and imaginative images.

From users:

“I really love the ability to create larger images natively without hires fix. No additional faces or limbs at 704×1024 so far!”

“Very very good, this is my favorite checkpoint so far!”



RPG model card is dedicated to Role Playing Game (RPG) portraits reminiscent of iconic titles like Baldur’s Gate, Dungeon and Dragon, Icewindale, and other contemporary RPG styles. The development of this model involved multiple iterations of merging packs combined with Dreambooth Training, resulting in an ongoing work in progress.

From users:

“First off, thank you so much for the wonderful model! It is by far my favorite one available at the moment and I hope you continue to develop it and add additional training to it.”

“Best SD model I’ve ever tried, hands down. The quality is amazing, so much fine detail in the textures of skin and clothing, and I’m even getting pretty good hands with this one. ;-)”

“Really beautiful model!”

Feature image credits.



Digital Artist

I’m a digital artist who is passionate about anime and manga art. My true artist journey pretty much started with CTRL+Z. When I experienced that and the limitless color choices and the number of tools I could use with art software, I was sold. Drawing digital anime art is the thing that makes me happy among eating cheeseburgers in between veggie meals.

More Posts

Contact and Feedback

Thank You!

Thank you for visiting the page! If you want to build your next creative product business, I suggest you check out Kittl!