CapCut’s new AI design studio and Seedream 5.0 Pro elevate browser-based creative workflows

CapCut unveils a revamped online Design Studio featuring an AI-assisted canvas, reusable Skills system, and advanced image editing powered by Seedream 5.0 Pro, aiming to democratise professional-grade design within a browser.

CapCut has updated its web-based Design Studio with a redesigned infinite canvas, a new Skills system and deeper use of ByteDance’s Seedream 5.0 Pro image model, in a move the company says is aimed at making more advanced design work available inside a browser. The firm said the latest version is intended to let users generate, edit and iterate on assets without moving between separate tools, while keeping the workflow open to people without formal design training.

The centrepiece is an AI-assisted canvas that CapCut says can be used for creation and editing in place, rather than through a separate prompt window. The company describes features such as natural-language creation, point-and-select editing, batch text replacement, background removal and outpainting, along with tools for comparing multiple versions side by side. The related product material also positions the canvas as a response to the growing use of flexible, layout-based design work in tools such as Figma, but with generation and editing built into the same space.

CapCut is also introducing what it calls Skills, a system for packaging repeatable workflows into reusable actions. The company says this includes preset routines for common jobs such as brand kits and social graphics, as well as custom Skills that users can save and share across projects. It has also outlined future additions, including a Skills library with auto-suggestions, tighter brand-kit connections and more consistent style control across outputs.

The most technically significant part of the update is the integration of Seedream 5.0 Pro, which CapCut says brings more precise image editing, layer separation and native multilingual text rendering. The company’s own explanations of the model say it is designed for more production-oriented output, with support for localised edits, layered assets and image generation in 14 languages. CapCut says the model is being rolled out to users from today, alongside the post-editing tools, while the Skills system is already live on the web.

Disclaimer: This content is intended for informational purposes only. Readers are advised to exercise their own judgement, conduct due diligence, or consult a qualified expert before acting on any information provided.