The popularity of AI has led to a host of websites and services that are able to create an image from a text description or prompt. One option worth trying is Stable Diffusion.
An effective and versatile tool, Stable Diffusion is available through a range of websites and as a standalone application. Simply type your description, and it generates corresponding images. Here's how it works.
Using the various Stable Diffusion websites
As a text-to-image model, Stable Diffusion is accessible through various websites.
Stable Diffusion Online
1. Generate the image
Use your browser to go theStable Diffusion Online siteand click the button that saysGet started for free. At the field forEnter your prompt, type a description of the image you want to generate. Then click theGenerate imagebutton. In response, the site displays four images by default.
2. Select an image
Click one of the four generated pictures and you can view it in a larger size. You can switch between the four images by clicking its thumbnail. Right-click on an image to display your browser's menu, which allows you to save, copy, or email the image.
3. Generate a new image
If you're not happy with any of the suggested images, just try again. Keep the same prompt, but click the Generate image button to see four different images based on your description.
1. Generate an image
Another website that uses the Stable Diffusion model is Hugging Face. Browse to the site and type your description at the prompt.
You should notice that there's also a field for Negative prompt. Here, you're able to type text and suggest anything you don't want to include in the image. Click the Generate Image button, and four images appear in response.
2. Select an image
Click a specific image to view it. Right-click on the image to save or copy it.
DreamStudio is another site that uses the Stable Diffusion model. However, this time you'll need to sign in with a supported account. You should also be aware that DreamStudio offers you a certain number of free credits that count toward generating images. If you use up those credits, you'll have to pay for more to continue using the site.
Also: How to use Bing Image Creator (and why it's better than DALL-E 2)
Browse to the Stability.AI login page. You can either log in with an existing Google or Discord account or click the sign up link at the bottom to create a new account. After signing in with your account, you're taken to the DreamStudio website where you can log in.
Before you can generate an image, you'll want to select a couple of options. First, click the field for Model and choose the version of Stable Diffusion you want to use -- either one of the production versions or the latest beta. Next, click the Style field and choose the type of style you'd like for your image, such as Anime, Photographic, Comic Book, Cinematic, or 3D Model.
Also: This new technology could blow away GPT-4 and everything like it
Under Settings, move the first slider to set the size of the image. Then move the second slider to choose how many images you wish to see.
1. Generate an image
Type your text description in the prompt. You can also exclude certain items from the image by adding them in the Negative prompt. When ready, click the Dream button. In response, the site displays the number of images that you selected in Settings.
2. Select an image
Click an image to see a larger version. Right-click the image to save or copy it.
Download and install Stable Diffusion
Rather than relying on the different websites that use Stable Diffusion as a model, you can install the program on your computer. However, be forewarned that setting up the application is challenging and requires several precise steps. The good news is that, once you're done, you can run it whenever and how often you want.
Also: The best AI art generators to try
Before you get started, you'll need to make sure your PC meets the minimum requirements: Windows 10 or 11, and a discrete NVIDIA video card with 4GB or more of VRAM. You can check the name and model of your video card and the amount of VRAM via the DirectX Diagnostic Tool. Press Win key+R to trigger the Run box. In the Open field, type dxdiag. In the DirectX Diagnostic Tool window, click the Display tab to see the name of your card and the amount of VRAM.
If your card is up to snuff, go through the following steps:
1. Install Python
Browse to the Python 3.10.6 website. Scroll down to the Files section and click the link for the Windows installer (64-bit) (assuming you're running 64-bit Windows). Run the downloaded file to install Python.
2. Install Git
Browse to the Download page for Git for Windows. Click the link for 64-bit Git for Windows Setup. Install the downloaded file, accepting all the default options.
3. Download the Stable Diffusion project file
Browse to the GitHub page for the Stable Diffusion project file. Click the green Code button in the upper right and select Download ZIP. Extract the downloaded ZIP file. Doing so creates a folder named stable-diffusion-webui-master. Copy this folder with all the extracted files to a simple location, such as the root of your C: drive.
4. Download and set up the checkpoint file
Browse to the Hugging Face webpage for the Stable Diffusion checkpoint file. Click the download link to save the 768-v-ema.ckpt file. This is a huge file, so be patient while it downloads. After the file has downloaded, navigate to the following location: C:\stable-diffusion-webui-master\models\Stable-diffusion (assuming you copied this folder to the root of your C: drive). Inside this folder is a file named Put Stable Diffusion checkpoints here.txt. Copy the 768-v-ema.ckpt file into this folder.
5. Download the config yaml file
Browse to the page for the config yaml file. The file will appear as straight text. Right-click anywhere on the page and select Save as. Navigate to the C:\stable-diffusion-webui-master\models\Stable-diffusion folder and save the yaml file in this location. Change the name of the file to 768-v-ema but keep the yaml extension, so it should be named 768-v-ema.yaml.
6. Run the webui-user.bat file
Move to the stable-diffusion-webui-master folder and run the webui-user.bat file. After a couple of minutes, the window should display a line that reads: Running on local URL, followed by a URL with the IP address of your local machine and a port number. Copy and paste that entire URL with the port number into your favorite browser to display the Stable Diffusion application. However, keep the command window open.
7. Generate an image
Now you can finally generate an image. In the Stable Diffusion application, type a description for the image you want at the prompt. Enter any text you wish to exclude in the Negative prompt. If you want to, you can change the parameters for Sampling method, width, height, and the rest. Set values for Batch count and Batch size if you need to change the defaults. When ready, click the Generate button. The application displays the images based on the count you requested.
Peek around the Stable Diffusion application and you'll find a host of other options, including ones to upload an image to generate a variation of it, scale and otherwise tweak an image, and tap into different extensions.
If you find an image you like, click it to enlarge it. Click the Save or Zip button to save it. Otherwise, all the generated images are automatically saved in a folder that you can access by clicking the Folder button.
More on AI tools
- How to use ChatGPT to build your resume
- How to use Bing Image Creator (and why it's better than DALL-E 2)
- How to use ChatGPT: Everything you need to know
- How to use ChatGPT to write code
- How to use ChatGPT to write Excel formulas
One of the most popular text-to-art AI generators right now, Stable Diffusion, is an open-source AI art generator that takes text prompts and outputs images in mere seconds.How do you get realistic images in Stable Diffusion? ›
- Negative prompt.
- Lighting keywords.
- Camera keywords.
- Facial details.
- Controlling faces. Blending two names. Blending one name. Inpainting faces.
- Generate images from text. The most basic usage of Stable Diffusion is text-to-image (txt2img). ...
- Generate an image from another image. Image-to-image (img2img) transforms one image to another using Stable Diffusion. ...
- Photo Editing. ...
- 4. Make videos.
One of the most popular text-to-art AI generators right now, Stable Diffusion, is an open-source AI art generator that takes text prompts and outputs images in mere seconds.How much GPU is needed for Stable Diffusion? ›
Stable Diffusion is not compatible with smartphones or most laptops. However, it can operate smoothly on an average gaming PC in 2023, provided that certain requirements are met. These requirements include a GPU with a minimum of 6GB VRAM, which is found in most modern NVIDIA GPUs.What image size is best for Stable Diffusion? ›
By default, Stable Diffusion generates images in sizes 512 to 512 pixels. You will get the most consistent result when you use this size. You can change the size, but it will require more computational power.Is Stable Diffusion as good as Dall E? ›
Stable Diffusion is more powerful
For all its ease of use, DALL·E 2 doesn't give you a lot of options. You can generate images from a prompt, and…that's kind of it. If you don't like the results, you have to tweak the prompt and try again.
|AI image generator||Price||Output speed|
|Bing Image Creator||Free||Fast|
|DALL-E 2 by OpenAI||Free + Credits (depends on sign up date)||Fast|
|Dream by WOMBO||Free + Subscription||Fast|
- Choose the Highest Resolution Output. ...
- Upscale Your Image. ...
- Use Outpainting to Expand Your AI Artwork. ...
- Fix Mistakes Using Photoshop. ...
- Download the Image Without a Watermark. ...
- Write a Good Prompt. ...
- Learn About Different Art Styles. ...
- Collect Inspiration for Prompts.
This means that stable diffusion can be used to improve product quality and consistency. On the downside, stable diffusion can be difficult to control, which can make it challenging to harness for practical applications. One example of stable diffusion is in the creation of diffusion models.
Stable Diffusion and Midjourney are two of the most exciting image-generating AI text-to-image models available today. They make it super easy to create great-looking artwork from just a few text prompts.Can you make money with Stable Diffusion? ›
There are several ways to earn an income with Stable Diffusion AI Art: Create and Sell Your Own AI Art: Create your own AI art, sell it on art marketplaces, and build up a portfolio of work. Consider selling on platforms like ArtStation, Artfinder, and Saatchi Art.What is the controversy with Stable Diffusion? ›
Stability AI, the startup that makes the popular AI art tool Stable Diffusion, faces two lawsuits. They allege the company infringes on copyrights by scraping the web to train its art algorithms.Is Stable Diffusion worse than Midjourney? ›
Easy to generate good images
If fact, very often, it will ignore part of your prompt and deliver surprising aesthetic images. A Stable Diffusion user needs to put more work into building a good prompt and experiment with models to generate an image of similar quality. Verdict: Midjourney wins.
Midjourney produced the best-quality images in our examples. They were more realistic and more beautifully rendered than DALL-E 2 and Stable Diffusion.Is 3090 better than 4090 Stable Diffusion? ›
The RTX 4090 is now 72% faster than the 3090 Ti without xformers, and a whopping 134% faster with xformers. The 4080 also beats the 3090 Ti by 55%/18% with/without xformers. The 4070 Ti interestingly was 22% slower than the 3090 Ti without xformers, but 20% faster with xformers.How much RAM do I need for Stable Diffusion? ›
You can definitely run Stable Diffusion with just 8GB of RAM, but having 16GB will undoubtedly result in a much smoother experience.Should I use CPU or GPU for Stable Diffusion? ›
Therefore, for Stable Diffusion, it is best to have a GPU. Though there are a few ways to run it without a GPU, they aren't as reliable as they may seem. So, make sure to have a good graphics card before running Stable Diffusion for the best results.Can you make NSFW images with Stable Diffusion? ›
The NSFW ("Safety") Checker
The Stable Diffusion image generation models will produce sexual imagery if deliberately prompted, and will occasionally produce such images when this is not intended. Such images are colloquially known as "Not Safe for Work" (NSFW).
Stable Diffusion Reimagine is a new Clipdrop tool that allows users to generate multiple variations of a single image without limits. No need for complex prompts: Users can simply upload an image into the algorithm to create as many variations as they want.
Image AI upscaler like ESRGAN is an indispensable tool to improve the quality of images generated by Stable Diffusion. In fact, it is so commonly used that many Stable Diffusion GUI has built-in support for it.Can Stable Diffusion generate 3D models? ›
TextMesh: A New Stable Diffusion-Based Text-to-3D Model From Google. TextMesh is a new text-to-3D work from Google that improves the now fashionable approach of using Stable Diffusion to generate different angles of the same basic prompt (2D picture) and then a 3D mesh is assembled from this using NeRF.What are the competitors of Stable Diffusion? ›
Stable Diffusion's competitors include Dream, PicSo, Booth AI, Freeway ML.What are the best AI diffusion models? ›
Blended diffusion and unCLIP are two prominent examples of such models that can generate highly relevant and accurate images based on user input. Also, GLIDE by OpenAI is another widely known solution released in 2021 that produces photorealistic images using user input.Which AI makes photos look better? ›
Enhance Image Quality With AI
Fotor's AI photo editor comes with a powerful photo enhancer tool that lets you improve photo quality online in an instant. Just upload your photo, and Fotor will automatically analyze it and fix colors and lighting issues, enhance sharpness, and remove blur from your images in real time.
The rendered images are created by a software called Lensa which was created by Prisma AI.How to make money with AI image generator? ›
Print-on-demand products made from your AI art are one of the most lucrative ways to make money. You should simply transfer your plans to sites like TeeSpring and RedBubble. Your artwork will be printed on a variety of products from mugs and t-shirts to posters and notebooks by these websites.Can AI generate 4K images? ›
An AI image upscaler is a type of software or hardware that uses Artificial Intelligence based algorithms to upscale image from a lower resolution to 4K one. The goal of the image upscaler is to upscale the quality of the image, making it sharper, clearer, and more detailed.Does AI-generated art sell? ›
Yes, AI-generated art can be sold just like any other artwork. In fact, there is a growing market for AI art, with some pieces selling for significant amounts of money. Here is a summary of the most popular styles of AI art which can be sold online.Can I legally use AI-generated art? ›
US Copyright Office: AI Generated Works Are Not Eligible for Copyright.
Cost of Stable Diffusion
As of now, Stable Diffusion is now available for free. It is a proprietary technology developed by OpenAI, a research organization specializing in artificial intelligence. To access Stable Diffusion, you must become an OpenAI partner and obtain a license.
In all this rage, you should take advantage of the open-source generative model, Stable Diffusion. It creates photorealistic images based on text prompts, similar to everyone's favourites – Dall-E 2 and Midjourney, but will be forever free.Is Stable Diffusion 2.0 better? ›
Thanks to this new text encoder, Stable Diffusion 2.0 can generate significantly better images compared to version 1.0, according to Stability AI. The model can generate images with resolutions of 512x512 and 769x768 pixels, which are then upscaled to 2048x2048 pixels by an upscaler diffusion model that is also new.How much does Stable Diffusion cost per image? ›
Q: How much does Stable Diffusion cost? Pricing for Stable Diffusion starts at $0.2 per image.Is Stable Diffusion ethical? ›
Ultimately, the debate over the ethical implications of stable diffusion AI is ongoing. While proponents argue that it can help to improve decision-making processes, opponents argue that it is ethically unjustified. It is up to each individual to decide which side of the debate they believe is more valid.How long should Stable Diffusion take? ›
when starting stable diffusion, or when changing the model, it takes a long time, from 15 to 20 mins.Can you sell Stable Diffusion images? ›
Yes, selling Dall-E art is not an issue. All users can fully commercialize their artwork generated using Dall-E. This even includes reprinting, selling, and merchandising those images. In addition, the images they generated during the research preview are also eligible.Does Stable Diffusion steal artwork? ›
A group of artists are suing the creators of image generators Stable Diffusion and Midjourney, The Verge reports, for using their art to train AI that steals their jobs. Stable Diffusion makes it trivially easy to knock off a particular artist's style.Is Stable Diffusion stealing images? ›
The lawsuit, announced on Saturday, claims that the Stable Diffusion tool used by Stability AI, DeviantArt, and Midjourney was trained on billions of copyrighted images scraped from the internet and contained within the LAION-5B dataset, which were downloaded and used by the companies “without compensation or consent ...What are the lawsuits against Stability AI? ›
In February Getty sued Stability for a “brazen infringement” of its intellectual property, stating that it believed the AI company copied more than 12m copyrighted images — and their associated metadata — without permission.
Stable Diffusion is a stunning tool for creating beautiful, lifelike images. While the tool has some potential downsides, the benefits outweigh any negatives. With its ease of use, flexibility, and stunning results, Stable Diffusion is a must-have for anyone who loves beautiful images.What is the limitation of Stable Diffusion? ›
While stable diffusion has several advantages in image editing, it also has its limitations: It can be computationally intensive and time-consuming, especially when dealing with large images or videos. The quality of the results may vary depending on the input data and the network parameters used.How many people use Stable Diffusion? ›
Then, there's Stability AI's open-source image generation model Stable Diffusion, which has been used on pop music videos, Hollywood movies and by more than 10 million people on a daily basis.What are the weaknesses of Midjourney? ›
Limitations of Midjourney
To put it another way, while Midjourney is capable of producing images that are extremely precise and detailed, it is unable to produce images that are highly imaginative or complex — images that go above and beyond what is conceivable in the real world.
Upscaling – Midjourney Wins
Midjourney differs from Dall-E 2 when you want to adjust images in variations and upscaling. OpenAI's art generator provides only variations. In Midjourney, you can use suffixes in your prompt to create further variations.
- Use different expressions and angles.
- Use different backgrounds if you can.
- Use a square or 1:1 ratio setting. By default, Stable Diffusion's default image size is 512 x 512 pixels, so using square images makes your input more similar to your desired output.
Upload the image: You can now upload the image that you have selected as the source for Stable Diffusion. Once uploaded, the system will use the image to create a new image based on the prompt.What is Stable Diffusion for image to image translation? ›
Stable Diffusion is a deep learning model released in 2022. It is primarily used to generate detailed images conditioned on text descriptions, inpainting (adding features to an existing image), outpainting (removing features from an existing image), and generating image-to-image translations guided by a text prompt.How hard is it to train Stable Diffusion? ›
Stable Diffusion is available for various uses in different models and ways. However, the main purpose is the same, which is to generate images. Therefore, you can use the training model to train Stable Diffusion for your personalized image content. It is easy and does not require much time either.How many photos do you need to train an Stable Diffusion? ›
5-10 images are enough, but for styles you may get better results if you have 20-100 examples. Many of the recommendations for training DreamBooth also apply to LoRA. The training images can be JPGs or PNGs.
Stable diffusion refers to the process of creating and distributing stable digital images that can be used for commercial purposes without the risk of theft or piracy. This technology has become popular in recent years due to its ability to protect digital artwork and make it more accessible to a wider audience.Can Stable Diffusion edit images? ›
With the recent boom in Stable Diffusion, generating images from text-based prompts has become quite common. Image2Image technique follows this approach which allows us to edit images through masking and inpainting. With rapid advancements in this technology, we no longer need masks to edit images.How long does Stable Diffusion take to generate an image? ›
Stable diffusion has a better balance between speed and quality and can generate images within seconds, while Disco Diffusion usually takes minutes (5–20 mins depending on GPU spec, image size, and parameters) Stable diffusion is stronger on image coherence.Is Stable Diffusion illegal? ›
Usage and controversy
Stable Diffusion claims no rights on generated images and freely gives users the rights of usage to any generated images from the model provided that the image content is not illegal or harmful to individuals.