How Can We Help?

We build GSD, a foundational model that generates video, voice, lip sync, effects, music and on-screen text at once, in sync. On it we build Studio, Supercut, Compose, Spark, Stream and an API.

The fourth architectural pillar of generative AI. It plans the whole video once on a world graph, then renders every layer in one pass.

The editing engine built into GSD. It changes one layer for the seconds you choose and leaves everything else untouched.

Spark for an explainer from a link. Compose to write once and publish everywhere. Studio for long-form video. Supercut to edit any layer on the web, iOS or Android. Stream to watch and share. The API to build GSD into your product.

150 s and beyond in one pass (5× SOTA Video Models), with no camera cuts, and up to 300 s on NVIDIA Blackwell B300. Spark explainers run up to five minutes.

Voice and lip sync in 35+ languages.

Wishtales is in private alpha. Join the waitlist; access opens in cohorts, with credits to create.

Basic is $30 a month, Creator $60 and Max $300. Spark is $2 per minute, up to five minutes, so $10 at most per explainer. The API is $2 per minute with ten edits included. See pricing.

It is designed not to. Harmful concepts and impersonation are left out of the world graph by design, so they are not generated.

Our Privacy Policy sets out what we collect, how we use it and how to delete your account.

You can permanently delete your Wishtales AI account and its data at any time. Delete my account.