Stay updated with our daily and weekly newsletters for the latest insights and exclusive content on leading AI developments.
If you’ve tried popular text-to-image AI art generators like Midjourney or the various versions of OpenAI’s DALL-E, you’ve probably noticed they all share a similar chat interface. Typically, there’s a space for you to input your text prompts, and the application responds with an image embedded in the message.
While this interface is user-friendly for many, some believe it’s restrictive, especially for professional artists and designers. Enter Visual Electric, a San Francisco-based startup offering a different approach to AI and visual creativity. Visual Electric aims to cater better to the needs of graphic designers and concept artists with more workflow-specific optimizations.
The startup, which emerged from stealth mode today after raising an undisclosed amount from Sequoia, BoxGroup, and Designer Fund, believes its approach is superior to the traditional chat-based interaction with AI models. Colin Dunn, founder and CEO of Visual Electric, explains that the company focuses on creating a better product experience rather than developing its own AI image generator. Instead, they are building on the open-source Stable Diffusion XL model, with flexibility to incorporate other models if they offer better results.
Unlike traditional AI art generators, Visual Electric provides an infinite virtual canvas where users can generate and move images around freely, rather than scrolling through a linear feed. This nonlinear approach aligns more closely with the creative process, allowing users to explore different ideas and revisit previous ones without restriction.
To assist users in generating high-quality images, Visual Electric offers a dropdown field with autocomplete suggestions for prompts, based on early user data. Users can also enter custom prompts. Additionally, the platform includes preset styles that mimic traditional art techniques like “marker,” “classic animation,” and “stained glass,” among others, with new styles regularly added.
Users can specify image aspect ratios and dominant colors via buttons and a sidebar, enhancing functionality similar to Adobe’s Firefly 2 AI art interface. Visual Electric also offers tools to “remix” images based on the initial prompt and “touch up” specific parts of an image using a digital brush, allowing for precise modifications.
Moreover, the platform includes an upscaler for enhancing image resolution and detail. Dunn highlights that these tools represent what they see as an AI-native workflow, designed in the order they’re typically used.
Although Visual Electric launches publicly today, it has been in alpha testing with a select group of designers who have provided valuable feedback. One client, for example, found the platform particularly useful for creating visual content for hundreds of universities efficiently and creatively, a task that previously required extensive manual editing of stock imagery.
An additional feature is Visual Electric’s “Inspiration” feed, a collection of AI-generated images from other users. This feed allows users to see prompts used and to remix images into their private canvas, similar to platforms like Pinterest.
Privacy is a significant concern for enterprise users, and Visual Electric ensures that private images are safeguarded, with more stringent privacy measures available in the “Pro” plan.
Visual Electric is launching with a tiered pricing model. A free plan offers 40 generations per day for personal use, while a standard plan at $20 per month ($16/month annually) provides unlimited generations and commercial usage rights. The pro plan at $60 per month ($48/month annually) offers higher resolution images and private generations.
Visual Electric aims to transform the creative workflow with its innovative platform, making the process more intuitive, flexible, and suited to professional needs.