Support

Vinetas uses open-source AI Models that run entirely on Apple Silicon via the MLX framework — the same technology that powers many leading on-device AI applications on Mac and iPad. When you tap Generate, the selected AI Model processes your prompt locally, producing image panels directly on your device. No request leaves your machine; there is no server, no queue, and no round-trip latency. The AI Models are downloaded once from a CDN and stored on your device — after that, generation works fully offline.

The number refers to the unified memory the AI Model requires to run. Bigger AI Models have more parameters, which translates to higher visual quality, better prompt faithfulness, and richer detail — especially in complex scenes. The 8 GB AI Model is fast and capable on any 8 GB+ Apple Silicon machine and is fully included in the Free tier. The 16 GB AI Model delivers a clear step up in quality — sharper, more prompt-faithful output suited to pitch decks, pre-vis, and production-grade storyboards — and is available on the Pro tier. Both AI Models support every preset style, custom prompt, and aspect ratio in the app.

Vinetas detects your device's unified memory automatically and highlights the AI Models that fit. The Settings screen shows each AI Model alongside its RAM requirement and current download status, so you always know what's available on your hardware. As a rule of thumb: any Mac or iPad with 8 GB of unified memory can run the 8 GB AI Model; 16 GB (or more) machines can run both the 8 GB and 16 GB AI Models. The app will never let you select an AI Model your hardware cannot support.

Yes. In addition to the 18 genre preset styles, Vinetas includes a Custom mode where you can write any style description you like. Custom prompts are available on both the Free and Pro tiers, on both AI Models, and support every aspect ratio and inference control the presets do. There is no limit on how many custom prompts you can use or save.

Every image you generate is saved to an on-device gallery inside the Vinetas app. Generations are stored in your app's private container and are never uploaded to any server. The gallery is searchable and retains the full metadata for each panel — prompt, style, aspect ratio, seed, steps, and guidance values — so you can reproduce or refine any result. You can also browse and manage your generations at any time from the Gallery tab.

Tap or click the Share button on any panel in the gallery to export it as a full-resolution PNG. PNG export is available on both the Free and Pro tiers with no watermark and no quality reduction. On macOS you can also drag panels directly from the gallery into other apps, or use the standard Save As dialog to choose your destination folder.

No. Your prompts and panels never leave your device. All generation happens locally on Apple Silicon — there are no cloud round-trips for image generation, no telemetry of what you create, and no logging of your prompts or storyboard content. The only network activity Vinetas performs is downloading AI Models from a CDN the first time you select them; after the download completes the app runs entirely offline. Standard Apple crash reporting (via App Store / TestFlight) applies if you have opted in to share diagnostics with Apple.

No. Not now. Not ever. Vinetas does not charge per generation, impose a monthly quota, or meter API calls — because there is no API. Everything runs locally on your device. You pay for access to bigger AI Models (Pro tier), not for the right to generate. Both Free and Pro are unlimited; the only practical limit is your device's available disk space for storing images.