Introducing 3D Jutsu
Higgsfield's new 3D workspace turns a prompt into an editable scene with objects, cameras and lighting, then renders it to video.
Higgsfield's new 3D Jutsu is an AI-native 3D workspace that works together with Supercomputer. A prompt or reference image turns into an editable 3D scene, with objects, layout, lighting, cameras and animation, which can then be rendered into video. Individual objects in the scene can be edited without regenerating everything, and shots can be organised and turned into finished video in one shared workspace.
To build a scene, a prompt is described and a model picked, including GPT-6 Astra, which produces a draft scene. From there, characters and objects can be moved, cameras placed, and lighting adjusted, working together with the AI agent.
