Scade Pro

The AI workflow builder is the center of the product. You drag nodes onto a canvas, connect them in order, and each node runs its own model or tool. Nodes show execution time and credit cost, so you can see which step is expensive before you scale a workflow up.

Scade Pro works as an AI app builder out of the box, with templates for common jobs, including sketch to image, virtual fitting rooms, competitor analysis, and shorts generation. Each template opens as an editable flow, which makes it a starting point rather than a locked product. You can publish a finished flow as an AI app with its own interface.

Agents take a flow and run it on a schedule or a trigger. The faceless video agent, for example, produces short clips for social channels without filming. This is the part that turns a one-off experiment into a repeatable process. You still need to check the output before publishing.

You can load your own documents into a private knowledge base and have models answer from that material instead of general training data. Scade also supports LoRA fine-tuning, which adapts an existing model to your style or dataset without training one from scratch. Both features are aimed at teams that need output tied to their own content.

Every published flow can be executed through a REST endpoint. You create a server key in the API and credentials section, then send a request with the start and end node IDs. Connectors and triggers push results into web apps, messengers, and e-commerce platforms, so the flow can sit behind a product instead of only inside the Scade dashboard.

Every action costs credits, and one credit is roughly one cent. A ChatGPT node task averages about 0.5 credits, while an image generation with Flux Pro costs around 6. The platform shows cost per node, which helps you compare a cheap text step against an expensive media step before you commit.