>_FAQ

Simsala is a local desktop app for visual AI workflows. You collect project images, references, CAD exports, prompts, and generated variants on a canvas and store everything in a local project folder.

Not quite. Simsala is more like a local working studio for design variants: import sources, set references, prepare annotations or crops, render variants, compare results, and keep the history of a project traceable on the canvas.

A local project is a vault folder on your computer. Simsala creates a fixed structure inside it for imported files, references, generated results, canvas states, and connector imports.

Project files, canvas states, imported media, references, and generated results are stored in the local vault. Simsala does not need a central server or cloud database for that.

Only when you start an AI render, the prompt and the selected image, reference, or video data are sent to the chosen AI provider. Simsala asks for external AI consent for this. Local project folders are not uploaded as a whole.

Yes. You can create a project, import files, organize the canvas, manage references, and collect connector imports. For image, upscale, or video renderings, you need a connected provider account.

Simsala currently supports Replicate and Runware. Depending on the model, both providers can be used for image, upscaling, and video workflows.

In Simsala, it means that you use your own account and your own API key at a supported provider such as Replicate or Runware. Simsala is not built as a credit reseller and currently does not provide a general interface for arbitrary custom provider endpoints.

No. AI costs are billed directly by your provider. Simsala shows provider and cost information in the app, but your Replicate or Runware account remains the billing account.

API keys are stored in the operating system secure storage, for example in the macOS Keychain or Windows Credential Manager. Simsala does not permanently store new provider keys in normal localStorage; old fallback entries are cleaned up on startup.

Yes. You can switch Replicate or Runware in the settings and add the matching API keys. Model options and render modes depend on the currently selected provider.

Yes. A Simsala license is intended for the productive render flow. The license is activated through Polar and checked before render jobs.

No. The Simsala license unlocks the app or the render flow. Provider costs, API limits, and credits run separately through your Replicate or Runware account.

The license key and activation data are managed locally and, where possible, stored in the operating system secure storage. The app also uses local caches so the interface does not depend on a fresh online check for every click.

You can continue opening and organizing local project content. AI renderings require internet access to the selected provider. After successful validation, license checks have a local offline grace period of 7 days.

Yes. Because a project is a local folder, you can back up or move the vault. When you open it again, Simsala selects the folder as the project. If you use cloud sync folders, avoid editing the same vault in parallel on multiple devices.

Connectors for Rhino, SketchUp, Vectorworks, and Archicad export views as connector packages. Simsala reads these packages from the local connector-inbox folder of the active vault and places them as canvas nodes. The convenient direct handoff runs through a local bridge on 127.0.0.1.

Simsala itself does not operate its own training backend for your project content. When you start renderings, the terms of the selected AI provider apply to the transmitted data.

Yes. The app can export a local diagnostic package. This is triggered deliberately through the settings and is meant for support or bug reports; it is not an automatic project upload.

Simsala is built as a Tauri desktop app for macOS, Windows, and Linux. Individual CAD connectors and installers are platform-dependent.