Leonardo is one of the most experienced photo editing platforms, having launched shortly after the consumer AI revolution began in late 2022. What sets this Australian service apart from its competitors is the wide range of features and high-quality output offered to free and paid subscribers. This is paired with a winning mix of tools for specific tasks ranging from marketing and social media materials to graphic and interior design. While the tools aren’t that different from other platforms, the presentation and ease of use make them stand out from the crowd.
Leonardo has a great user interface and is easy to use on a daily basis. While products like Canva and Adobe Firefly have the edge in terms of user numbers and reputation, Leonardo is definitely a product to keep an eye on in the coming months and years.
I looked at the whole package and tried out as many features as possible to find out if this versatility results in a loss of speed or image quality.
Leonardo review: First impressions
You can set up an account with one of the usual suspects, Apple, Google, Microsoft or via email and when you sign up you have a choice of different plans from the free plan at $0 to the Apprentice plan at $12 per month and Maestro at $60. The differences are in the speed of generation, the number of monthly tokens and the number of your own models you can train every 30 days. The free plan offers 150 tokens worth of image generation per day. That’s about 10 to 20 images.
Once you’re up and running, you’ll be presented with a homepage that’s busier than a Tokyo train station on a Friday night. Every inch of the site is filled with a variety of services, tools, guides, and user examples. This isn’t a bad thing, but it can be a little overwhelming at first. But take a deep breath and dive in, because that’s when the fun begins.
Leonardo test report: In use
Basically, Leonardo is just an AI image generator and editor, but the options you get with this simple list are numerous. For example, you can generate an image from the dashboard on the home page, doodle an image and let the AI create a correct version, generate an instant image from your words while you type and also edit images, upscale them, and convert them to videos. It’s exhausting just trying to choose something you can do.
My tip for the most fun is to dive into the “Realtime Gen” menu option at the top of the page. Then click into the prompt box below and prepare for a mind-boggling effect. Type in some words and the image will appear as you type. The longer and more detailed the prompt, the more complex and/or accurate the image will be. It’s insane. Then you can start playing around with the element and genre settings on the page to customize the result. Fabulous. This isn’t the only example of instant photos on the market, but it’s definitely one of the most fun and versatile.
The default image generation option is also extremely powerful. Midjourney may take the crown in image coherence and beauty, but Leonardo is way ahead of anything else on the market in terms of overall performance and flexibility. The Ideogram service, for example, places a lot of emphasis on its text processing capabilities, but Leonardo manages to handle text too with its new Leonardo Phoenix model. Of course, much of this is thanks to the underlying Stable Diffusion 3 upgrades, but it just goes to show how far image generators have come in just a few months.
I have a prompt for a “Street photo of a 19 year old urban grunge punk skater, grubby clothes, leaning against a wall, raining, wearing a t-shirt with the words ‘Rock The Kasbah’ on the front.”
The platform’s prompt enhancer improved the prompt to “A 19-year-old punk skater exudes urban grunge vibes in a rainy alley, leaning against a weathered brick wall. His grubby clothes perfectly match his rebellious spirit, and a tattered t-shirt boldly reads “Rock The Kasbah.” This striking image, perhaps a photograph, captures the young man’s nervous presence in exquisite detail. The rain-soaked sidewalk reflects the neon glow of nearby street lamps, giving the scene a moody atmosphere. Every little detail, from his unkempt hair to the scuffed skateboard beside him, is rendered with cinematic precision, immersing the viewer in a gritty urban narrative.”
The results were great and were generated within 30 seconds. Of course there were a few minor things that could be tweaked, which is normal. The text quality was excellent, as was the promptness of compliance, although some of the resulting images had a rather odd pose. AI is clearly a bit of a contortionist.
But that’s where the Canvas Editor comes in. This is probably one of the most comprehensive editors on any online AI image platform. It allows you to modify the model, paint it in or out, change the canvas dimensions, and delete elements on an image. You can really go wild and customize the resulting image to your liking.
You’re always limited to the possibilities of the image model you choose, but that’s not a problem because there are hundreds of Leonardo and community models to choose from. These include the traditional options like anime and photography, but also some beautiful, offbeat models like K-pop idols and 3D travel figures.
To use the editor, all you have to do is load an image and then select the tools you need. I found the abundance of options quite daunting at first, but with a little patience and using the excellent Leonardo Help Guides I was able to get some good results from my testing. Painting over was pleasingly easy, as were the principles behind using Focus mode to edit small sections of an image.
However, using Inpaint to fix some generation errors proved to be quite more difficult than it should have been. No matter what settings I chose to edit or delete small parts, the results were almost always disappointing. Leonardo is not the only application that suffers from this problem. Stable Diffusion models often fail these tests, presumably because the model has trouble reading the image context correctly. I think there is room for general improvement.
A similar problem occurred with the Motion feature, which supposedly creates an animation from a still image. It still seems to be a gamble, and my results were almost always either comical or a complete failure. I suspect this simply reflects the early state of the art (pun intended) of Image2Video technology at the moment. It’s usually not bad for a simple camera pan across the stands, but it’s painful for anything else.
One of the most useful options for business users in the Leonardo toolkit is the Training & Datasets feature. This allows you to optimise your own brand dataset by using images from your own marketing assets. This allows you to recreate your brand identity on demand – perfect if you need to create new social media, advertising or marketing content.
Unfortunately, unless you already have a strong brand presence or want to establish a strong product style, it won’t help much, as the fine-tuned mockups currently can’t provide cohesive text alongside the images. So basically, you might as well use Leonardo Phoenix’s standard mockup to create more generic shots of your shoes, pizza slices, or coffee, and add your own design elements later with Canva. But fine-tuning is an interesting tool to look out for in the future as technology catches up.
Leonardo test: Conclusion
I really like Leonardo. It’s not a mind-blowing product yet, as there are still a few issues with image generation and processing floating around the edges of the platform. But when it hits the mark, it can definitely hold its own against anything on the market. One of my standard tests for any AI generator is to create a “beautiful tiger,” and so far only Midjourney has managed to truly pull that off in one go. But Leonardo does that too. It created a majestic tiger head portrait in the desired setting, and I was amazed.
Aside from this accuracy, the additional tools are the icing on the cake. It’s extremely handy to have a range of options in one interface, especially if you’re a busy professional. Some of these, like motion, texturing and fine-tuning, are clearly a work in progress, but the others are really useful. The real-time editing, upscaling and overpainting all work brilliantly and the selection of models is spectacular.
All in all, this is a solid, professional AI image generation platform that is constantly improving. It’s easy to use, flexible, and offers surprisingly good value for money considering what else is on offer. Oh, and did I mention it’s fast? I look forward to seeing what comes next as the general backend technologies become more mature and developed.
