Creating images from a simple text prompt is one of the most well established uses for generative artificial intelligence and there are dozens of AI image generators on the market offering an equally wide range of options, features and styles.
In less than two years we’ve gone from tools like Midjourney being able to create a low-resolution, barely recognizable depiction of a human to high definition, photorealistic images you can barely distinguish from those taken with a camera.
We also now have inpainting, consistent character and upscaling tools from StabilityAI, well utilized by companies like Leonardo and NightCafe, as well as text on images from OpenAI in DALL-E 3 and Ideogram, the AI startup from former Google engineers.
I spend most of my day during the week and even at the weekend using one or more AI image generators, pushing them to the limit, seeing what they can achieve and how easy they are to use. These are the best AI image generators you can access right now and every one on this list something slightly different or works in a different way.
Best AI image generator overall
Leonardo is at the core a very well done wrapper for a variety of Stable Diffusion models much like a number of others in the same category but it takes it so much further. With custom styles and fine-tuned versions of models as well as other AI image tools — it is a stand out in the field.
Its ability to generate photorealistic images, thanks to the fine tuned PhotoReal model, is nearly up to Midjourney standards and it can create a range of styles with the Elements feature.
These elements are a fine-tuned model that can be applied before generation and steer the image to be created with a particularly look such as a sketch or sculpture. You also have the ability set a style such as cinematic, food or long exposure.
What really makes Leonardo standout for me is the combination of an easy to use UI, mixed with an incredible amount of control. You can add reference images and set how the AI should use them, control the size and layout, even add a transparent background.
Most of these features are available in other platforms, but Leonardo has them all as well as a range of other features such as image upscaling, live image generation and one of the most creative tools — the ability to draw a sketch and have the AI turn it into a full image.
Best for Photorealism
Stubonly sitting inside a Discord server, Midjourney is one of the most high profile and impressive AI image generators currently available. It fails on several counts including ease of use but what makes it harder to use also makes it more impressive.
It is particularly good at creating photorealistic images, with some of the more talented users able to get it to create pictures that look like they’re straight from a phone camera. Midjourney was one of the first to crack the finger problem and has consistently real looking people.
Midjourney is somewhat controversial for its refusal to discuss the source of its training data. Many suspect much of it comes from scraping any publicly available images it could find regardless of whether it had permission from the image creators.
What makes Midjourney really stand out for me though is the level of control you have over every aspect of the generation. You can use parameter commands to reference the style or a character within another images or use others to completely change how an image appears.
The latest version six update also comes with the ability to add legible text to images and create hyper realistic product images, although this isn’t consistent or always reliable.
Best for text on images
Ideogram is one of my favorite AI image generators for my own personal use. It isn’t the best in terms of feature set but it follows a prompt really well and can add text like no other model. I’ve been able to generate full movie posters, flyers and greeting cards with accurate text.
Accessed through a wonderfuly simple prompt box, with the option to automatically enhance your prompt to get a better image, it is both easy to use and powerful.
While it is best for adding text on to images, it also has a bit of Midjourney flare to the style of the works it generates. You can turn off magic prompt and create more artistically simple images, or even add customized style tags.
Ideogram's Magic Prompt is a compelling feature. If its turned on then a large language model will analyze your prompt and re-write to be much more descriptive to get closer to your vision.
You can see your original prompt as well as the magic prompt for any image, adapt it or use it to create a new picture. You can also use any generated image as a source for a new image.
Best for creativity
Some image generators are completely standalone like Midjourney and others are built into another product such as Microsoft's Designer, which forms part of the Copilot chatbot. It is also available for free without paying for Copilot Pro.
Built on the same underlying DALL-E 3 model used in ChatGPT, Microsoft has created something genuinely impressive with Designer. It lets you customize any aspect of the image, even pulling out individual elements within the picture.
You can make some subtle changes within the chat UI, or you can edit in Designer, opening it the full Microsoft image editor. This goes beyond simple AI changes and allows you to change the background, add filters, text or other visuals.
One of my favorite features is color pop. You can select any one or more objects within the generated image, click color pop and it will make the background more greyscale.
As well as making changes within the Designer interface, such as changing the aspect ratio or giving it a new style, you can work within the Copilot chat to add elements or make other more substantial changes. This could include changing the clothes of a character or a car type.
Best for interaction
DALL-E 3 is only available in ChatGPT to those with a Plus account. There are a few ways to use DALL-E within ChatGPT. You can just access it through the main interface, through the DALL-E GPT custom chatbot, or by tagging DALL-E in the main chat.
The original DALL-E was one of the first high profile commercial generative AI image tools. Initially available as an API or through a dedicated DALL-E page, OpenAI has since bundled it with its chatbot. This is also its primary selling point — the ability to talk through an image.
Everything is based on text prompts and it uses completely natural language for generation. For example you can tell it to generate a picture of a cat, then follow up with asking it to add a hat.
With the most recent update you can now click on the image itself and make edits within the picture. This is done by drawing over the part you want to change and telling ChatGPT how to change it — once again relying on the conversational nature of the editor.
I don’t think DALL-E is the best AI image generator but it is a good all rounder. It can do text, it can create photorealistic images (with a slight uncanny valley) and produce artistic work but the ability to reason and rationalize over the image with text is the big benefit.
Most innovative
Google’s Imagen 2 AI image generation model is one of the best around. It creates engaging and original images and can handle text on images as well as Ideogram. There are a few ways to access it but the most innovative is the ImageFX experiment from Google Labs.
What makes ImageFX fascinating is the way it handles prompting. You give it your paragraph or so long prompt and it picks out specific keywords, turning them into dropdown menus. Each menu then gets three or four alternatives similar to the word you used.
For example if you ask for a picture of a gorilla in glasses giving a lecture while wearing a suit it might tag up suit, gorilla, glasses and lecture. You could then, at the touch of a button, switch out glasses for sunglasses or lecture for driving lesson.
While this is only an experiment and the same images can be foud in Google Gemini (which didn’t make my list), the flexibility and novel approach to prompting gave it the win.
The biggest downside to ImageFX is it can only make square images — the same problem Meta’s Imagine and Google Gemini have. Most offer a range of orientations but the fun way of prompting the model, quality of images and rapid generation from ImageFX make up for it.
Best for ethical training
Adobe Firefly has some impressive tools including prompt suggestions, deep customizations for the image generation and a training dataset almost exclusively trained on Adobe Stock pictures.
This last point means it has a more ethical training set than most image generators on the market, even prompting Adobe to offer financial indemnity from copyright claims against images generated using Firefly. There is also a second generation Firefly coming soon.
I do find that Firefly isn’t as good at creating photorealistic images as Midjourney or Ideogram but its artistic skills are among the best. It also produces compelling graphics, all of which makes sense given the more artistic nature of the Adobe Stock library.
Adobe offers a number of generative AI features including vector generation, template creation and generative fill in Photoshop, all powered by the Firefly model.
One of the best features of Firefly is also one of the most recent. It is called Structural Reference and lets you carry the layout of one image over to another.
More from Tom's Guide
- OpenAI's GPT Store is now live with over 3 million custom chatbots to try
- OpenAI custom chatbots — this is how they work and what they can do
- Forget ChatGPT — OpenAI now lets you build your own chatbot for any task
