Midjourney
Midjourney, Inc.Text-to-image generator known for stylised, art-directed output
by Leonardo Interactive (Canva)
Image and video generation workspace with fine-tuned models and asset tools
Leonardo AI is a generation workspace aimed at people producing large volumes of visual assets, particularly for games, apps and social content. Alongside text-to-image it offers a canvas editor, an upscaler, background removal, texture generation for 3D models, real-time sketch-to-image, and a motion feature that animates stills into short clips. A catalogue of community and in-house fine-tuned models lets you pick a specific look rather than prompting your way toward one, and element weights adjust style strength. Since the company was acquired by Canva the product has continued as a standalone tool with its own account and pricing. It suits indie game developers, content teams and freelancers who value throughput and organisation over one perfect image. The limitation is that the interface carries a lot of settings, model choices and credit types, and newcomers often spend their first session working out which combination applies to what they want.
Create an account at leonardo.ai; a daily allowance of free tokens refreshes each day, which is enough to learn the tool. The home screen pushes you into image generation, where the important decision is the model you select from the dropdown, because that governs the style far more than your wording does. Set the number of images low while you experiment, since each generation spends tokens and the count multiplies. Once you have a base image, the canvas editor handles inpainting and outpainting, and the universal upscaler produces print-ready sizes. For animation, send a still into the motion module and expect short clips rather than full scenes. The main setup issue is token accounting: fast generation, upscaling, motion and premium models draw at different rates, and a free daily pool disappears in minutes if you leave high settings on. Subscriptions add a monthly token pool, private generation and relaxed queueing that does not consume the fast allowance.
Text-to-image generator known for stylised, art-directed output
Real-time image and video generation with live canvas control
Community hub for open image models, LoRAs and shared prompts
Generates editable vector graphics, icons and illustrations in a brand style