StabilityAI shocked the world with its release of Stable Diffusion in August 2022. It allowed anyone to input text and generate images using AI and output it to various styles - creating infinite possibilities.

Shortly after Stable Diffusion v1.4 and v1.5 were released, users started training their own custom models on top of the base models. These new, more focused models, allow users to fine-tune certain styles and emphasize nuances of a style (some even able to perform better than stable diffusion's base models).

We've been playing around with Stable Diffusion models and compiled the Top 10 Best Models to use to generate stunning images with zero code experience or downloads necessary.

Where to find Stable Diffusion Models?

All the latest models can be found at either Civitai.com or HuggingFace.co.

Both generally have the same models, however, HuggingFace does seem to crack down on licensing more, so you'll generally find that more problematic (NSFW) models are located on Civitai. Civitai also comes with an easier user interface so I recommend starting with that out of convenience.

Tip: many of the below websites will require account creation due to NSFW content

How to use the models

The fastest way to test out the models is to use one of these sites. They require you to create an account, and as of posting, none require a payment method.

On account creation, you'll receive free credits and be able to use test out any models you like.

• Tensor Art (Use for any model besides SDXL 1.0 - results are subpar)
• Clip Drop (Recommended for SDXL 1.0)
• Sink In
• Dazzle AI
• Mage Space (Currently only allows testing on SD 1.5, 2.1)
• Random Seed

Once you've created your account, you can select the model that you want, and input the prompts just like you would in stable diffusion web UI.

If you don't know which to pick, start with Tensor Art (has generous free credit allotment that refreshes daily with no credit card required).

Top 10 Best Models

1. ChilloutMix

Explore the Model

Best For: Kpop Idol look, Photorealism

This model remains one of the most downloaded on Civitai and aims to recreate the Korean Doll look that is popular amongst Kpop Idols today. The model is capable of both sfw and nsfw outputs and is slightly limited in the facial variation due to the similarity of images used in the training set.

Example Test Prompt

Negative Prompt

2. Deliberate

Explore the Model

Best For: Generating perfectly specific images, Art designers

Provides the ability to create the exact image that you are thinking of. This model does better with more detailed prompting. Allows you to describe as much as you'd like and will reward more meticulous descriptions.

Example Test Prompt

Negative Prompt

3. DreamShapr

Explore the Model

Best For: Illustrations, Fantasy Art, Characters

This started as a model to make good portraits that do not look like cg or photos with heavy filters, but more like actual paintings. The result is a model capable of doing portraits, but also great backgrounds and anime-style characters.

Example Test Prompt

Negative Prompt

4. NeverEnding Dream

Explore the Model

Best For: Cosplay Characters, Anime Pictures, Good-Looking People

A model that is "artistic" like DreamShaper but also able to use LoRA's and booru tags for generating images. This model is also able to make good anime stuff out of the box, without needing additional LoRA's.

Example Test Prompt

Negative Prompt

5. Realistic Vision

Explore the Model

Best For: Photorealism, Portraits

This model is great for pictures that look like they were taken by a camera. Able to be used with textual inversions that can model celebrities and well-known figures as subjects in the photos.

Example Test Prompt

Negative Prompt

6. MajicMix Realistic

Explore the Model

Best For: Photorealism, Good-Looking People

This model aims to be the realistic version of the MajicMix model. Through training, this model is able to recreate the camera effects, skin texture, and lighting to appear more photorealistic.

Example Test Prompt

Negative Prompt

7. Hipoly 3D Model LoRA

Explore the Model

Best For: Video Game Characters, 3D Models

This is a LoRA trained on high-polygon 3D model images.

It can provide clean, high-resolution skin and hair materials, as well as detailed clothing elements.

- Model Creator

Example Test Prompt

Negative Prompt

8. GuoFeng

Explore the Model

Best For: GuoFeng Style Characters

GuoFeng, in Chinese, roughly translates to "National Style", and it has become a broad online trend among influencers. It refers to the traditional dress of the Han people and this model aims to recreate that style.

Example Test Prompt

Negative Prompt

9. CyberRealistic

Explore the Model

Best For: Photorealism, Portraits

Introducing my versatile photorealistic model - the result of a rigorous testing process that blends various models to achieve the desired output. While I cannot recall all of the individual components used in its creation, I am immensely satisfied with the end result. This model incorporates several custom elements, adding an extra layer of uniqueness to its output.

One of the model's key strengths lies in its ability to effectively process textual inversions and LORA, providing accurate and detailed outputs. Additionally, the model requires minimal prompts, making it incredibly user-friendly and accessible.

- Model Creator

Example Test Prompt

Negative Prompt

10. A-Zovya RPG Artist Tools

Explore the Model

Best For: Concept Art, Book Characters

Made specifically with RPG artists in mind. Great for generating concept characters as well as image styles that are most used in board games and book covers.

Example Test Prompt

Negative Prompt