Shared controls across all tabs

Prompt bar

The prompt bar sits below the generation tabs and is available in every tab. It contains the primary text input, AI writing aids, and the Generate button.

Instructions / Prompt
The main multi-line text field. Describe the space, material, edit, or scene you want to generate. The prompt combines with any active tab settings and pill selections to produce the final instruction sent to the AI.
Prompt idea chips
Up to four short suggestion chips appear below the prompt field. These are context-aware - they change based on the active tab and selected space type. Clicking a chip fills the prompt box with that suggestion; clicking it again clears the field.
✦ Enhance
Rewrites the current prompt in rich, professional design language - expanding material descriptions, lighting, atmosphere, and stylistic nuance. Replaces the prompt text in place. Its code-level flat fee is shown under Credits & pricing. Disabled if the prompt field is empty. In the Video tab, Enhance uses Veo cinematography guidelines instead.
◎ Describe
Analyses the currently selected canvas image and writes a detailed prompt describing its style, materials, lighting, and spatial mood. The generated text fills the prompt field automatically. Its code-level flat fee is shown under Credits & pricing. Requires an image node to be selected on the canvas first.
◆ Generate
Sends the prompt, tab settings, and any source image to the AI and produces a new result node on the canvas. Up to 5 generations can run concurrently - the button becomes disabled at that limit and shows the count. Disabled states: no prompt text and no active pill template, no canvas image selected (for image-to-image tabs), or the concurrent limit reached.
▶ Generate (Video)
When the Video tab is active, a dedicated video Generate button replaces the standard one. Requires the video prompt field to contain text. Generation via Veo 3.1 takes 1–3 minutes.
Upscale button
When the Upscale tab is active, an Upscale button replaces Generate. Requires a source image (dropped into the drop zone or selected on canvas), a model, and a scale multiplier selected.
Vector mode buttons
In Vector mode, the button changes with what's selected: with no source image it reads Generate and creates new vector artwork from your prompt; with a Selected Image it instead offers Vectorize or Remove BG as one-click actions in place of a prompt-driven generation.