Visual Electric launches an AI-powered image generator with a designer workflow focus

Visual Electric, a institution backed by Sequoia Capital, has launched its generative AI-based image-generation instrumentality aimed astatine designers.

The institution was founded successful November 2022 by CEO Colin Dunn, a erstwhile executive astatine Universe, a mobile building tract (Dunn besides worked astatine Facebook and Dropbox); Chief Product Officer Adam Menges, a erstwhile Apple worker whose startups Lobe (a creation tool) and Lowkey (an AR crippled company) were acquired by Microsoft and Niantic, respectively; and Chief Technology Officer Zach Stiggelbout, who worked astatine Microsoft connected Lobe.

Dunn told TechCrunch complete a telephone that a batch of generative AI-based image creation interfaces are not focused connected nan workflows of designers.

“We started Visual Electric because generative AI is an interface problem. Existing devices trust connected a linear chat metaphor. But productivity isn’t linear, it’s messy. So we built our merchandise astir nan imaginative process. Our generative canvas lets you dispersed out, make a messiness and activity spatially,” he said.

The interface is elemental to get started: You person to type successful a punctual to get 4 options based connected an infinite canvas. You tin iterate arsenic galore times arsenic you want to make a suitable image for your project. While penning a prompt, you tin besides prime a colour palate, what colors to exclude, format and a temper including Airbrush, Film, Cinematic and Neon.

The instrumentality is free to usage but location is simply a headdress of 40 image generations per day. You tin salary for nan premium scheme that starts astatine $16 per period for amended procreation speeds and nary limit connected image creation on pinch a licence to usage these pictures commercially. Visual Electric said that nether some these plans, immoderate image you make is shared pinch nan community. However, if you salary for nan top-tier $48 per period plan, your snaps are backstage and you get moreover amended velocity for image generation.

The instrumentality besides has features for illustration “touch” to modify different parts of a generated image pinch a prompt, an upscaling characteristic for premium users, nan expertise to region nan inheritance aliases make variations by changing nan “creativity” slider and a “remix” characteristic to springiness an existing image a makeover.

The institution was founded successful 2022 and it has raised $2.5 cardinal successful backing from Sequoia, BoxGroup and Designer Fund.

Visual Electric uses Stable Diffusion arsenic nan image procreation exemplary on pinch GPT-3.5 Turbo to powerfulness auto-generated punctual suggestions. The instrumentality besides deploys GPT-4 to make a punctual from an imported image.

The institution doesn’t person a mobile app yet but is considering moving connected it adjacent year.

In nan past year, companies for illustration Google, Meta, Amazon, Adobe and moreover Getty Images person launched galore image-generation tools. Chatbot platforms for illustration ChatGPT and Poe person besides integrated image generations for easier access.

Visual Electric thinks that its attraction connected a circumstantial group of users will springiness it an edge. Recently, different devices for illustration Vercel and TLDraw — which precocious raised an further $2 cardinal successful funding — person besides experimented pinch interface-generating features.