Brief IA

Runway Unveils Solaris, No-Code AI Interface for Real-Time Rendering

🎨 Creative AI·Tom Levy·

Runway Unveils Solaris, No-Code AI Interface for Real-Time Rendering

Runway Unveils Solaris, No-Code AI Interface for Real-Time Rendering
Key Takeaways
1Runway unveils Solaris, an AI model that generates no-code interactive interfaces
2Solaris combines reasoning and rendering to produce real-time environments
3User interactions, whether gestures or text, modify the interface live, including virtual try-ons
4No availability date has been announced
💡Why it mattersSolaris illustrates the emergence of world models, which could transform the creation of digital interfaces without requiring programming skills.
Le brief IA que lisent les pros

Le brief IA que les pros lisent chaque soir

Les 7 actus IA du jour, décryptées en 5 min. Gratuit.

Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.

Choisis ton rythme

Gratuit · Pas de spam · Désabonnement en 1 clic

📄
Full Analysis

Runway unveils Solaris, an AI model capable of generating live interactive interfaces and applications without advanced programming skills. Mouse or keyboard interactions and text commands drive the evolution of the screen, extending to uses like virtual try-ons. The model is part of the rise of "world models," although its commercial availability has yet to be specified.

No release date announced and a model aligned with the "world" wave

Runway has not communicated a release date for Solaris and did not immediately respond to a request for comment. Solaris fits into the emerging trend of world models, touted as the next major evolution in AI. These models aim to surpass the capabilities of traditional language models by seeking to translate and transform aspects of the physical world. Solaris is cited as one of the first examples of this new generation.

Direct interactions and text commands, extending to virtual try-ons

Solaris takes into account user actions, such as clicking or dragging an element, to generate new images in real time. It also accepts text instructions, such as changing the color of an object or moving an element within an image. Runway illustrates these possibilities with an example of virtual try-on: the user can drag a shirt from a rack onto their own image to see the result. The entire system operates like a streaming video that continuously adapts to interactions.

Reasoning and rendering combined for an image-by-image generated interface

The operation of Solaris relies on the combination of reasoning and rendering. The model generates the interface image by image, while a language model determines the evolution of that interface. This combination allows for the creation of real-time digital environments that respond to user actions.

A no-code world interface model for creating websites and apps

Runway has presented Solaris as a world interface model. This system enables the design of interactive websites and other types of interfaces by directly manipulating the elements displayed on the screen. Solaris does not require advanced programming skills and relies on live rendering.

Brief IA — L'actualité IA en français

L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.