AI Video Reel | Pipe2.ai

Back to pipelines

Bring your next idea to life

Start with a prompt or try an example.

Complete to generate

Video Segments

Upload reference videos

Up to 10 short clips

Browse library

Creative Direction(optional)

Describe the story, audience, pacing, order, visual style, audio balance, or moments that matter.

Examples

Before / After

A YC interview, cut for social

A one-minute answer from a Y Combinator interview becomes a 37-second vertical clip with a hook title, word-by-word captions, punch-ins and tighter pacing. The question and the full answer stay intact. Source: Y Combinator on YouTube.

Reproduce

0105

How to use Video Reel

A compact decision guide for placing this pipeline in a workflow.

Best for

AI editing of 1-10 source clips into one finished, social-ready video
Talking videos: tight pacing, word-timed captions, a hook title, punch-ins and inserts from the sources
Clips without speech: a montage cut on the music beat with corrected exposure
Adding supplied or generated music, narration or a generated voiceover

Tips

Describe the story, audience, order, pacing and moments that matter
Say which moment should open the video; the first second decides whether people keep watching
For clips without speech, add or request music so the cuts can follow its beat
To preserve product colors, say so explicitly; exposure is corrected and color is graded only when requested
Use Edit Style Preserve for a natural edit without captions, zooms or added titles

Video Reel API

Call this pipeline from your own code. One request dispatches a run; the model is an input field, not a separate endpoint.

Get an API token

inputs: 15
required: 1
Pricing: 0.1 to 11 credits

Language

TypeScript Python Go CLI

bun add @pipe2-ai/sdk

Run a pipeline

import { createClient } from "@pipe2-ai/sdk";

const client = createClient(process.env.PIPE2_TOKEN!);

const { run_pipeline } = await client.RunPipeline({
  pipeline_slug: "video-reel",
  input: {
    video_urls: ["a1b2c3d4-0000-4000-8000-000000000001","a1b2c3d4-0000-4000-8000-000000000002"],
  },
});

pip install pipe2

Run a pipeline

import asyncio
import os

from pipe2 import Pipe2Client

client = Pipe2Client(token=os.environ["PIPE2_TOKEN"])

run = asyncio.run(client.run_pipeline(
    pipeline_slug="video-reel",
    input={
        "video_urls": ["a1b2c3d4-0000-4000-8000-000000000001","a1b2c3d4-0000-4000-8000-000000000002"],
    },
))

go get github.com/pipe2-ai/sdk-go

Run a pipeline

client := pipe2.NewClient(os.Getenv("PIPE2_TOKEN"))

input := json.RawMessage(`{
  "video_urls": ["a1b2c3d4-0000-4000-8000-000000000001","a1b2c3d4-0000-4000-8000-000000000002"]
}`)

run, err := pipe2.RunPipeline(context.Background(), client, "video-reel", input)
if err != nil {
    log.Fatal(err)
}

brew install pipe2-ai/tap/pipe2

Run a pipeline

pipe2 pipelines run \
  --pipeline video-reel \
  --input-json '{"video_urls": ["a1b2c3d4-0000-4000-8000-000000000001","a1b2c3d4-0000-4000-8000-000000000002"]}' \
  --wait

Per page

Input	Type	default	Description
`video_urls`required	array	default—	Upload or select 1-10 source videos. Their order is available to the editor as part of your brief.
`captions_mode`	enum	defaultauto	Auto follows your direction. Add burns verified speech captions into the finished video. None omits captions. auto · add · none `autoaddnone`
`captions_style`	string	default—	Optional language, placement and visual style for on-screen captions.
`edit_intensity`	enum	defaultmedium	For Bold and Clean: how tightly pauses are cut and how often framing, graphics and montage cuts change. low · medium · high `lowmediumhigh`
`edit_style`	enum	defaultauto	Auto gives talking videos a social style and cuts clips without speech to the beat. Bold adds punchy captions, zooms, titles and sound effects. Clean uses calmer captions and subtle zooms. Preserve keeps a natural edit. auto · bold · clean +1 `autoboldcleanpreserve`
`instructions`	string	default—	Describe the story, audience, pacing, order, visual style, audio balance, or moments that matter.
`music_direction`	string	default—	Optional mood, instruments and tempo for generated music.
`music_mode`	enum	defaultauto	Auto follows your direction. Add uses your uploaded track or generates music. None omits music. auto · add · none `autoaddnone`
`music_url`	string	default—	Upload or select an optional background music track for the reel.
`narration_url`	string	default—	Upload or select an optional narration track for the reel.
`preserve_source_audio`	boolean	defaulttrue	Keep useful audio embedded in source clips and mix it with optional narration or music.
`target_duration_sec`	number	default—	Optional maximum length. AI selects and paces useful moments to fit it.
`transition_mode`	enum	defaultauto	Auto follows your direction. Add uses short transitions between source shots. None uses straight cuts. Describe a preferred transition in Creative Direction; custom motion graphics are outside Video Reel. auto · add · none `autoaddnone`
`voiceover_mode`	enum	defaultauto	Auto follows your direction. Add layers a generated voiceover; Replace substitutes generated speech for source speech. None omits voiceover. auto · add · replace +1 `autoaddreplacenone`
`voiceover_script`	string	default—	Optional exact words to speak. Without a script, AI drafts only from verified source content.

Point any MCP client at pipe2 and your agent gets these tools. It finds this pipeline, reads the same schema above, then runs it.

list_pipelines
get_pipeline_schema
run_pipeline
get_pipeline_run_status
request_upload

https://mcp.pipe2.ai/mcp

{
  "mcpServers": {
    "pipe2ai": {
      "url": "https://mcp.pipe2.ai/mcp",
      "headers": { "Authorization": "Bearer YOUR_TOKEN" }
    }
  }
}

MCP setup

Full API reference CLI reference

Recipes using this pipeline

Show all recipes

Articles about this pipeline

Show all articles

Frequently Asked Questions

How many source clips can I provide?

Add 1 to 10 clips and create a reel up to 5 minutes long.

How does AI decide what to include?

It reviews the supplied clips and your creative direction, then chooses moments, cuts, pacing, framing, transitions, and audio that support the requested result.

Can I control the edit?

Yes. Set an optional maximum duration and describe the story, order, pacing, style, audio balance, or specific moments you want. Video Reel keeps the source format; use Video Reframe separately to change the aspect ratio.

Can I keep audio from the clips?

Yes. Enable Keep Clip Audio to retain useful dialogue and sound. It can be mixed with optional narration and background music.

How much does it cost?

Review the estimate shown before you start. The final charge depends on the work needed and stays within the displayed maximum.