storytold/artcraft

What it solves

ArtCraft provides artists with precise control over AI-generated imagery and video, moving beyond simple text prompting to a "crafting" workflow. It addresses the lack of consistency, repeatability, and intentionality in standard text-to-image generation by allowing users to layout scenes and pose characters before generating final assets.

How it works

It functions as an Integrated Development Environment (IDE) for visual creation, combining various AI models with traditional digital art tools. Users can compose scenes using 2D and 3D elements, perform character posing, and use image-to-location mapping to maintain environmental consistency. The platform integrates a wide array of third-party AI providers (such as Grok, Midjourney, and Sora) and its own internal models for image and video generation.

Who it’s for

Digital artists, filmmakers, and creators who need professional-grade control over AI assets and want to move from random generation to intentional scene composition.

Highlights

• 3D Scene Blocking: Use 3D asset kits and image-to-3D mesh conversion to precisely position elements and define camera angles.
• Character Control: Dynamic character posing and identity transfer using 3D mannequins as ControlNets.
• Advanced Compositing: Support for both 2D and 3D image compositing, including background removal and layer management.
• Multi-Model Integration: Access to a vast library of image and video models from various providers in one interface.
• Image-to-Video: Tools to transform static images into cinematic video with detailed controls.