Streaming answer

An assistant's answer streaming in: the prompt in a bubble, a moment of thinking dots, then the reply arriving word by word, each fading up, with a caret while it writes.

  • type/stream
  • Type
  • text
  • Full frame
  • 6 s (3–30 s)
Style
Live in your browser
Streaming answer at 16:9
0.00 / 6.00 s
npx volito render type/stream --style pop --out stream.mov
Playing live with volito/browser, the engine a render uses. Each tab lays the item out again for that shape, not a crop.

Render it

npx volito render type/stream --prop prompt="Cut this interview into a 30-second teaser" --out stream.mov

That writes ProRes 4444. The same item in another shape, brand or format:

volito render type/stream --aspect 9:16 --style cobalt --out stream-9x16.mov
volito render type/stream --format mp4 --out stream.mp4
volito still type/stream --t 3 --out stream.png
volito info type/stream     # every prop and part
volito add type/stream      # copy its source into ./motion to edit

From Node:

import { findItem, renderVideo } from "volito";

await renderVideo(findItem("type/stream"), { props: { prompt: "Cut this interview into a 30-second teaser" }, out: "stream.mov" });

Or ask an agent with the MCP server connected: “Render the Volito streaming answer (type/stream) in 9:16, in our brand.”

Props

Set each with --prop name=value, or all at once with --props and a JSON object.

PropTypeDefaultTakes
prompttextCut this interview into a 30-second teaser≤ 120 chars
response
Reply
longtextDone. I kept the three strongest answers, trimmed the pauses, put the hook first and timed the cuts to the music. Captions are on and the colour matches your last video.≤ 400 chars
speed
Words a second
number92 – 30

Parts

Each part takes any move, ease or idle: --motion ground.in=pop.

PartWhat it is
groundthe ground
promptthe prompt bubble

Tags: ai, chat, streaming, prompt, assistant

More in Type