Video Trim

From 0.1 credits

An AI video editor for source footage. Describe a fragment to find or material to remove, and get separate clips or one cleaned video.

Bring your next idea to life

Start with a prompt or try an example.

Complete to generate

Describe the fragment to find or material to remove, whether to return one cleaned video or separate clips, and any count or duration limit. Timestamps are optional. For example: find the part where two people discuss pricing; remove unambiguous filler words from the full video; or return three interesting moments, 10–15 seconds each.

Examples

Before / After

The leap and the catch

A 39-second clip of a border collie at play, trimmed to the run-up, the mid-air catch and the landing. The editor found the moment from a plain-language description, no timestamps needed.

Reproduce
0103

How to use Video Trim

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

Best for

  • Finding several interesting moments in a long recording
  • Describing a conversation, event or topic and receiving a matching clip
  • Removing unambiguous filler words, stutters, long pauses, or named sections while preserving the remaining speech
  • Creating one or more edited videos before reframe or captions

When to use

  • After importing a video when the user wants selected moments or unwanted material removed
  • Before video-reframe or captions

Tips

  • State whether you want separate clips or one cleaned video
  • Name exactly what to keep or remove and include count or duration constraints when relevant
  • Use the returned clips as inputs to Video Reframe or Captions

Video Trim 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
4
required
2
Pricing
0.1 to 11 credits
Language
Language
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-trim",
  input: {
    video_url: "a1b2c3d4-0000-4000-8000-000000000001",
    instructions: "Find and return the part where two people discuss pricing",
  },
});

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-trim",
    input={
        "video_url": "a1b2c3d4-0000-4000-8000-000000000001",
        "instructions": "Find and return the part where two people discuss pricing",
    },
))

go get github.com/pipe2-ai/sdk-go
Run a pipeline 
client := pipe2.NewClient(os.Getenv("PIPE2_TOKEN"))

input := json.RawMessage(`{
  "video_url": "a1b2c3d4-0000-4000-8000-000000000001",
  "instructions": "Find and return the part where two people discuss pricing"
}`)

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

brew install pipe2-ai/tap/pipe2
Run a pipeline 
pipe2 pipelines run \
  --pipeline video-trim \
  --input-json '{"video_url": "a1b2c3d4-0000-4000-8000-000000000001", "instructions": "Find and return the part where two people discuss pricing"}' \
  --wait

Full API reference CLI reference

Recipes using this pipeline

Show all recipes

Articles about this pipeline

Show all articles

Frequently Asked Questions

Can I ask for several clips?
Yes. Ask for a number in the instructions, such as three moments of 10–15 seconds. The result contains one library video for each selected moment.
Can I describe one moment instead?
Yes. Describe the event or conversation in plain language. The agent chooses a useful start and end and keeps the result within the duration limit.
Can I request an exact duration?
Yes. Say how long the clip should be, for example 20 seconds. The agent uses that as a target while keeping the moment understandable.
How long can each clip be?
Each clip is limited to two minutes. Use the optional maximum duration field for a stricter limit.
Does the agent inspect the complete video?
Yes. It reviews the source evidence before selecting moments, then renders and checks every returned clip.
What does this cost?
0.1 to 11 credits per run. The run can return several clips.

Video Trim: Find and Cut the Moments That Matter

Give Video Trim a long video and describe what you want. The AI watches the full source, finds one or more matching moments, and returns separate MP4 clips. Ask for three interesting moments, a conversation about a specific topic, or one exact event with a duration such as 20 seconds.

How It Works

  1. Understand the request: natural-language instructions describe the topic, number of clips and target duration
  2. Review the source: the agent analyses the complete video and checks the chosen ranges before rendering
  3. Return separate clips: each result is saved to your library and can be sent to Video Reframe or Captions

Describe the number, topic and target duration in the instructions. Video Trim chooses the ranges; it is the replacement for a separate highlights step.

Explore more pipelines

↑ Back to create See all →
Video Generator
from 9.5
Video Generator
Image Generator
0.2-6.7
Image Generator
Audio Generator
0.002-40.0
Audio Generator
Music Generator
2.5-22.6
Music Generator