Focused AI removal workflow

Remove a video object with a prompt

Upload a clip and describe the object you want removed. Medux uses the prompt to locate the target before processing the video, so be specific about its appearance or position.

Start with 3,000 free trial credits for eligible tasks. Compare free credits and usage limits.

Try it now

Object Removal by Prompt

Open in Playground
01

Choose a clip and describe the object

Upload a video, then name the unwanted object and where it appears. Use sample to see an example prompt, and replace it to match your clip.

No video selected
MP4, MOV, WebM, or AVI
Describe the target in your own video. The sample phrase is an example; replace it if it does not match your clip.

The latest task appears below. Processing continues on the server if you leave this page.

Use only media you own or are authorized to edit. Review the processed result before publishing it.

Result preview

Your result will appear here

No processed video yet. Generate one to see it here.

Build with Medux

From your browser to the API and MCP

Use the service behind this Workshop in your application or AI agent. Explore the API reference and its corresponding MCP tool before starting an integration.

Inspect the matching API task

Compare the request fields and response in Playground, then use API Docs for uploads, authentication, and task status.

Open the matching Playground task
One connected workflow

How it works

01

Upload the clip

Choose footage you own or are authorized to edit.

02

Describe the object

Write a concise prompt that identifies the target by appearance and position. No region box is required.

03

Generate and inspect

Medux uses the text to initialize object selection, then processes the video for review.

Creative control

Edit by describing what should disappear

Distracting objects

Name one object that draws attention away from the intended scene.

Quick creative review

Test a removal idea without drawing coordinates on a frame.

Agent workflows

Use the same input_file_id and prompt_text fields when orchestrating the API.

Browse all removal tools or inspect the matching Playground task.

Better results

Write a precise removal prompt

Name the target

Use a concrete noun such as bicycle, sign, or bag instead of a vague instruction.

Add location

Mention where the target appears if the video contains similar objects.

Review moving frames

Check the edited area after the target moves or passes in front of another subject.

Build with Medux

Run this tool from your agent or code

Every Medux tool has open examples on GitHub: copy a prompt for Claude or Codex, or start from a working API request.

REST API

API example

Request fields, curl, Python and JavaScript samples, and response handling for this endpoint.

View on GitHub
Frequently asked questions

Object Removal by Prompt FAQ

How is this different from selecting a region?

The prompt-based API takes input_file_id and required prompt_text. It uses your words to identify the object, while Video Object Remover takes a required array of pixel-coordinate regions.

Can I try Object Removal by Prompt with free credits?

New Medux users receive 3,000 trial credits for eligible tools. Processing tasks consume credits and plan limits apply. Check Pricing for the current limits and your Account for the final task cost.

Can I use media from someone else?

Use media you own or are authorized to edit. If a real person appears, make sure your use of the footage and likeness is permitted.

Where is my finished result?

Preview and download the completed result on this page. The submitted task is also available in Account Results.

Your next creation starts here

Try Object Removal by Prompt in one place

Try it now