The team behind Leonardo.Ai aims to stand out in the competitive AI image generation market by leveraging advanced natural language processing.
The origin of the image generation startup Leonardo.Ai’s name is no coincidence. CEO J.J. Fiasson explains, “Leonardo da Vinci embodied the fusion of artistry and innovation. I’ve always been captivated by the intersection of technology and creativity.”
In the bustling world of AI startups, image generation is a hotly contested arena. Leonardo.Ai competes with industry giants like OpenAI’s Dall-E 3 and Adobe’s Firefly, as well as platforms such as Midjourney, Pixlr, and Night Journey. All share a common goal: to simplify visual content creation to the point where it requires only typing or speaking a sentence.
Since its inception in October 2022, Leonardo.Ai set out to develop a generative AI platform that grants users greater mastery over image creation, harnessing a deeper understanding of everyday natural language.
The platform’s initial features included tools like style, content, and character references to accurately capture user intentions, as well as an AI canvas for editing. A company spokesperson states, “These functions enable users to employ natural language prompts while still achieving detailed, high-quality images, giving them full creative control over the entire process.”
The platform has reached 17 million registered users, with monthly active users ranging from 4 to 5 million in May.
Focus on precise prompts According to the spokesperson, the upcoming foundation model, Phoenix, offers exceptional control by “adhering to prompts with extreme accuracy.” Prompt engineering, or the art of communicating exactly what you want to a generative AI model, can be a delicate process on any platform.
Phoenix is the first model developed in-house by Leonardo.Ai. Currently in preview, its launch is scheduled for later this month.
Like others, the model can generate images based on “long and detailed prompts” with a “heightened level of prompt adherence,” the spokesperson said.
Prior to Phoenix, Leonardo utilized open-source models based on Stable Diffusion. These were trained on publicly licensed images, Creative Commons images, and synthetic data generated by algorithms or computer simulations.
The startup also allows users to train custom models to meet specific creative needs. As of December, 470,000 such models had been created.
In late 2023, Leonardo.Ai launched Realtime Canvas, a generative AI tool that enables real-time painting with AI assistance.
Another recent feature is Edit With AI, which allows image editing through text prompts.
The Realtime Gen function generates images as you type.
Fiasson explains, “It’s about simplifying the entire process. We want people to feel in control, rather than feeling like it’s a game of chance where they might or might not get what they want.”
To begin, users choose a preset model like Phoenix, enter a prompt, and click Generate to see an initial set of images.
When requesting an image of a kitten, teddy bear, and unicorn on the beach, the platform suggests modifying the prompt with “instructive language to quickly iterate,” such as “Add a red hat” or “Make it vintage.”
Users can also choose from presets like Anime, Graphic Design, and Stock Photography.
The platform even offers to enhance queries with adjective-rich prompts:
“A playful kitten, a cuddly teddy bear, and a majestic unicorn frolic together on the sandy shore beneath a bright sun. The kitten’s fur shimmers in the golden light, the teddy bear’s plush is pristine and inviting, and the unicorn’s horn gleams with magical energy. This enchanting scene is depicted in a stunning painting that captures the whimsical charm and innocence of these beloved creatures. The vibrant colors and intricate details make this image a true work of art that radiates joy and wonder.”
In her review of Leonardo.Ai, CNET’s Katelyn Chedraoui rated the tool as a solid option, comparable to competitors like Firefly. She noted that it indeed adheres to prompts better than platforms like Midjourney but lacks comprehensive editing tools and what she described as a “wow” factor.
Leonardo.Ai offers a free, limited version of the platform, along with premium plans starting at $10 per month. Features like character, style, and content reference require a premium plan, as does training custom models.
Initially targeted at the gaming industry, the platform has broadened its scope to include marketing and advertising, design, and entertainment, though Fiasson states that the tool is also intended for the general public. Currently, Leonardo.Ai is developing a business-focused offering, allowing enterprise users to train models using their own datasets and collaborate within the platform.
In December, Leonardo.Ai raised $31 million from investors including Blackbird, Side Stage Ventures, Smash Capital, Tirta Ventures, Gaorong Capital, and Samsung Next.
Fiasson concludes, “Much of our focus is on how to make the platform more powerful and controllable. How do we enable people to have a greater degree of control over the output? And how do we facilitate people using technology to move from an initial concept to a highly refined endpoint?”
This article is part of a series of brief profiles on AI startups, designed to help you understand the landscape of artificial intelligence and the activity in this field. For more information on AI, visit our AI Atlas hub, which includes product reviews, news, tips, and explanations.
New Gemini Gmail