Make long-form, on-brand videos with agents
Typecut builds the video from your idea or script — combining generative models, curated assets from across the web, and unique motion design into audio-synced scenes that look made, not generated.
For channels that publish on a cadence — uniformly branded, every timeMade with Typecut
Documentaries, history, explainers, faceless channels — stories that scale, rendered scene by scene.
Your script is the storyboard
Paste a script — or just an idea — and Typecut breaks it into scenes, writes the narration, and matches every beat with imagery and motion. No timeline. No keyframes.
Edit video at the speed of text
Scenes live inside your script. Reorder a sentence and the timeline follows. Swap a visual without re-rendering everything around it.
Never brief it twice
Configure a channel once — voice, pacing, fonts, caption style — and every project it produces starts already on-brand.
Everything between your idea and a finished video — research, script, voice, visuals, motion, and render — lives in one place.
Research & script
Give it a topic. Typecut researches the angle and writes the full narration.
Frontier narration
Ultra-realistic AI voiceover, consistent across every episode of a channel.
Sourced & generated visuals
Entity-grounded stock plus best-in-class image models, picked per scene.
Motion design
Typographic, cinematic scenes composed per beat — not a slideshow.
On-brand by default
Set a channel once — every project inherits its voice, pacing, and look.
Agent-ready
Drive it over MCP — agents write scripts and edit scenes.
Stop switching tabs. Start creating.
Research, script, voice, visuals, motion, captions, and render — one workspace, zero exports.
Built for every channel. Documentary, history, explainers, faceless uploads — Typecut gives one operator the whole production chain, tuned to the format.
Start creatingFrom script to screen
Frontier TTS voices. Ultra-realistic narration from Gemini, ElevenLabs and more.
Self-improving scripts. The writer learns viral structures and refines your narrative over time.
Latest image models. Seedream, GPT Image & Flux — best-in-class visual generation.
Zebra motion engine. A proprietary compositor for cinematic, typographic motion graphics.
Audio-reactive sync. Speech-synced animation with beat-reactive motion design.
Cloud rendering. GPU-accelerated exports in the cloud — no local hardware needed.
From solo creatives to autonomous pipelines
Or let an agent do it.
The same engine is exposed as a public API and MCP server. AI agents can write scripts, bring resources, and edit scenes — illustrating anything, at any length, and handing back a link you can jump into and refine.
POST /v1/projects { "script": "the-fall-of-rome.md", "preset": "documentary", "duration": "8m" } 201 Created { "id": "prj_8f3kq2", "status": "planning" }
Pricing
The Zebra motion-design engine on every render.
The long-form pipeline — 10+ minute videos from a single script.
GPU cloud rendering with unlimited watermark-free exports.
Frequently asked questions
All you need is a story to tell
Typecut handles the rest — script, voice, visuals, motion, and render.