Characteristic

Under the film

Everything you just watched is one editable program.

In Cadence, a decision isn’t a pixel. Every cut, ramp, grade, title and track is a line in a program: the thing the model writes, the editor opens, and the engine renders. Change a line and everything downstream re-flows.

How Cadence thinks

Five layers, one intention.

1. L1

   Intent

   What should this feel like, and for whom. The brief, your references, your characteristic.

2. L2

   Structure

   What happens, in what order. Takes, story beats, hooks, the shape of attention.

3. L3

   Treatment

   Time, color and sound as curves. Ramps, grades, beat sync, the mix.

4. L4

   Composition

   Type, graphics, masks and tracks. Hand-drawn motion and titles that sit inside the shot.

5. L5

   Scene

   Depth, light, cameras and CG. The frame stops being flat.

Underneath, a perception stack watches motion, meaning, speech and music. On top, your style conditions every layer, learned from a handful of your own edits.

Questions

What Cadence is, plainly.

• What is Cadence v1?

  Cadence v1 is the first model from Auteuriq: an AI model trained only for video editing, with its own native editing engine. It makes the decisions an editor makes, from which take to use and where to cut to colour, speed, titles, tracking and compositing, and it learns your style from a few of your own edits.

• What can Cadence v1 do?

  It reviews every clip for horizon, focus, exposure and motion, selects the strongest takes, assembles a story that opens on a hook, cuts on the beat of the music, ramps and freezes time, grades colour against live scopes, draws charts and hand-drawn motion, places 3D titles inside the shot, tracks moving objects, and builds shots in layers for compositing and CG.

• How does Cadence learn an editor’s style?

  From a handful of your finished edits it learns your characteristic: how long you hold a shot, how often you cut on the beat, which transitions and grades you reach for. It then brings that taste to new footage. It doesn’t learn your presets. It learns your eye.

• Who is Cadence v1 for?

  Creators, editors and studios who cut a lot of video and want to keep their own style while spending less time on the mechanics of editing.

• Can I use Cadence v1 today?

  Cadence v1 is in a private research preview. Auteuriq is opening it to a small group of creators, editors and studios first. You can request a seat with the early-access form on this page.

• Is the film on this page real footage?

  No. Characteristic is rendered live in your browser with WebGL, Web Audio and code, with no video files. Every edit in it was authored by the Auteuriq team to show where Cadence v1 is headed; it is not model output.

• What happens to my data?

  The characteristic in chapter 12 is computed on your device from how you scroll and never leaves your browser. If you join the early-access list, Auteuriq stores your email and role to contact you about Cadence. It is never sold or shared, and you can ask for it to be deleted at any time. Read the privacy notice.