Motion engine

Cinematic motion from a handful of stills.

This is the reason Etchly costs a fraction of generative video. Instead of paying a model to invent every second of footage, we generate a few illustrations and animate them with our own engine: depth, camera moves, reveals, characters and kinetic type.

Start free See pricing

Why stills plus motion beat per-second generation on cost

Generative video models such as Seedance, Veo, Kling and Sora bill by the generated second, typically between a few cents and about $0.75 a second through API providers in late 2026. Clips run 5–15 seconds, so a one-minute video is several generations stitched together, and any shot with a melted hand or a wrong prop has to be paid for again.

Etchly flips the math. A 60-second video usually needs around 16 illustrations at 1 credit each. Our engine supplies the movement, so the cost stays flat however long the camera lingers on a scene, and a re-render that reuses the same art is free.

About 16 illustrations for a 60-second video
Motion is rendered, not generated, so the price doesn't grow with every second of video
Re-renders that reuse art cost nothing

The Produce step showing generated illustrations and the rendered video

Depth, camera and parallax that feel shot, not slideshowed

Every illustration can be cut out into layers: subject, midground and background. The engine moves them at different speeds, so a slow push-in has real depth. Full-frame plates get Ken Burns-style pans and zooms timed to the narration, and cut-out characters can step forward from the scene.

Camera moves are directed per scene in the script, so a reveal lands on the exact word that calls for it and not on an arbitrary cut.

Multi-layer parallax from a single illustration
Push-ins, pans and drift timed to narration
Full-frame plates or cut-out subjects

Line art that draws itself, characters that talk

Line-art and engraving styles are vectorized, so the engine can etch them onto the screen stroke by stroke, the way an illustrator would. It's a signature look that generative video can't reliably produce, and it costs nothing extra.

For stories and dialogue, characters can talk with mouth movement synced to the voiceover, and on-screen titles, numbers and lists animate in with the words that trigger them.

Draw-on and etch animation for line art
Talking characters for dialogue scenes
Kinetic titles, numbers, lists and compare layouts

Rendered natively, about 10 seconds a minute

The engine is our own Canvas-based renderer, not a browser screen recorder. It draws every frame natively and encodes in parallel across render workers, so a 60-second 1080×1920 video renders in about ten seconds. The same engine powers the live preview in your browser, so what you scrub is what you download.

1080×1920 or 1920×1080 H.264 MP4
Frame-accurate, word-timed animation
Live preview uses the same engine as the final render

When you still want true generative video

We're honest about the trade-off. If a shot needs photoreal people moving, real-world physics or live-action footage, a generative model is the right tool. For explainers, stories, countdowns, listicles, product walk-throughs and most social ads, animated illustration looks deliberate and polished, and it costs a fraction as much. Read our guide to choosing.

Works with

Related features

Illustration styles

Line art, engraving, watercolor, neon, paper collage, risograph, 3D clay and cinematic scenes.

Learn more

AI review

An AI reviewer that watches the render and fixes timing, overlaps and clipped text.

Learn more

Captions & audio

Word-timed burned-in captions, original music and SFX, per-effect faders, −14 LUFS.

Learn more

FAQ

Motion engine: questions

Is the motion AI-generated?

The illustrations are AI-generated. The motion is rendered by Etchly's engine: parallax, camera moves, draw-on line art, character animation and kinetic titles. That's why you don't pay for every second of footage.

Will viewers think it's a slideshow?

No. Layers move independently for depth, the camera is directed per scene, and reveals land on the words that trigger them. It reads as an animated, produced video.

Can I control the camera?

Yes. Each scene's motion is part of the scene plan. Adjust it directly, or describe the change in plain English, like “slow push-in on the storefront”.

Make your first video free.

10 free credits on signup, enough to script and produce a short 15-second video. No card required.

Start free