Select a node that can generate and a generator bar appears at the bottom of the canvas. This is the console for one run: write the prompt, pick the model, tune parameters, read the estimate, hit send.

What the generator bar contains

Roughly, top to bottom, the bar has:

• The prompt box: describe the image / video / voice you want; type @ to reference connected nodes or items from your saved library (see References and @ mentions);
• Model select: models differ in capability and price; you always see the model's display name; with none picked, the default model for that task type is used;
• The parameter area: resolution, aspect, duration, count and so on, varying by model and task type; only parameters the model truly supports are shown;
• The reference strip: image-to-image, image-to-video and the like show thumbnails of the connected image / video / audio;
• The send button: submits the run, with the estimated credit cost shown live beside it.

Estimate and billing

Every time you change the prompt, model or a parameter, the bar re-quotes the server and refreshes the estimate, so the price you see and the basis you are actually charged on are identical.

• The estimate shows as one of "— / N credits / up to N credits": models billed per token, per minute or per 5 seconds show "up to", with the real amount settled when the task finishes;
• When a model is unpriced or misconfigured, the estimate slot shows the matching notice rather than a wrong number.

Submit and results

After you hit send, the node goes Queued, then Running, and the result lands back on the node when it finishes.

• You can cancel mid-run: cancelling does not refund spent credits, and the node reverts to its last successful result;
• On failure the system decides automatically whether to refund — only a "Refunded N credits" badge on the node means a refund actually happened;
• Double-clicking send never double-charges; there is concurrency dedupe protection.