Adobe launches Firefly, its own generative AI tool
Adobe has announced the launch of an artificial intelligence tool called Firefly that will let users type commands to quickly generate text and images, according to a blog post published Tuesday. Developed using hundreds of millions of photos, the first Firefly model is designed to generate images and text effects
Adobe has announced the launch of an artificial intelligence tool called Firefly that will let users type commands to quickly generate text and images, according to a blog post published Tuesday.
Developed using hundreds of millions of photos, the first Firefly model is designed to generate images and text effects from descriptions.
According to Adobe, Firefly will enable creators to use their own words to generate content the way they dream it up, from images, audio, vectors, videos and 3D to creative ingredients, like brushes, colour gradients and video transformations, with greater speed and ease than ever before.
A screenshot from Adobe's Firefly release video shows AI image manipulation based on a text prompt
The company says the model will soon be integrated into its industry-leading tools including Express, Photoshop, Illustrator and Adobe Experience Manager so users can effortlessly leverage the power of generative AI within their existing workflows.
On a technical level, the Firefly model shares similarities with text-to-image AI bots, such as OpenAI's DALL-E 2 and Stable Diffusion, as they all have the ability to transfer image styles and produce new images from textual descriptions. However, according to Adobe, Firefly intends to circumvent the ethical challenges associated with content duplication that has plagued many of its competitors.
The first Firefly AI model has now been launched in beta and the company says it will be focused on commercial use.
