Vitent Product Guide

Create a video with Vitent Agent

Start with a goal, a guided template, or the structure of an Inspiration case. Review every source asset and generation plan before creating the final video.

Updated September 10, 2026

Create a video with Agent mode

Who Agent mode is for

Agent mode is designed for people who need a useful video but do not want to begin by learning scripts, asset systems, storyboards, or node-based canvases. It translates a business or content goal into a guided production task.

Choose Agent mode on the Start Creating page — Select Agent on the left side of Start Creating. You can describe the goal directly or choose a template below the input box.

Three ways to start

Describe the result directly. State what you are promoting, who should watch, where the video will be published, and what action viewers should take.
Choose a template. Use a prepared workflow such as One-Click Reference, TV Commercial, Food & Beverage, or Vlog when the task already matches a common production pattern.
Start from an example. Browse Inspiration, find a result close to your goal, and use it as a reference for format, pacing, subject, or visual direction.

A useful first request usually includes the product or topic, target audience, publishing platform, desired duration, visual tone, available materials, and call to action. If information is missing, the Agent can ask for it before preparing the generation task.

Example: Create a 15-second vertical product video for Instagram Reels. The audience is first-time skincare buyers. Use the attached product photos, keep the tone calm and premium, and end with “Shop the launch collection.”

Complete a video with the Agent

Define the goal: choose Agent mode and describe the business or content outcome.
Select a starting structure: continue from the conversation, a template, or an example.
Provide materials: upload product images, clips, logos, reference videos, or other assets requested by the workflow.
Confirm the brief: review the audience, platform, duration, aspect ratio, style, message, and call to action.
Let the Agent prepare the task: the Agent organizes the required assets, prompts, shots, and generation steps.
Review and revise: comment on the result in plain language, keep what works, and request targeted changes.
Export and publish: confirm the final result and download it for the intended channel.

When an action consumes credits, Vitent shows the estimated amount before generation. Review the scope and cost on the confirmation card before proceeding.

Example walkthrough: remix a skincare commercial

This walkthrough was completed in the live English interface on September 10, 2026. It starts from the Inspiration example A Quiet Skincare Ritual and creates a new product video with replacement assets. The product and brand used below are demonstration assets generated for this test; they are not claims about a real brand.

A Quiet Skincare Ritual Inspiration case with Watch Now and Remix actions — Open the Inspiration case, review the visual direction, and select Remix.

The Agent loads the example's production structure: a 15-second, 16:9 skincare commercial with eight beats from product appearance to the closing brand frame. It does not reuse the example product as the new product. Instead, it asks for a clear product reference with the container, opening method, and texture.

Agent asks for a facial skincare product reference and offers Generate for Me — Upload your own product reference, or choose Generate for Me when you are testing the workflow without prepared materials.

For this test, Generate for Me created a fictional LUMIÈRE Hydrating Face Serum reference. The image approval card showed the exact prompt, model, dimensions, and an estimate of 15 credits before submission. The Agent then proposed a matching adult demonstrator reference for another 15 credits.

Approval card for generating the replacement skincare product reference — Read the generated prompt and live credit estimate before you confirm a new product asset.

Generated serum and adult demonstrator references in Agent project assets — The generated product and demonstrator are saved as project assets and used consistently in the screenplay and storyboard.

After the two references were ready, the Agent wrote the screenplay, produced the storyboard, and prepared one continuous 15-second video plan with eight internal cuts. The live plan used Wan 3.0, 720p, 16:9, audio enabled, and displayed an estimate of 110 credits before creation.

Agent approval card for one 15-second skincare video plan estimated at 110 credits — Review the bound references, shot instructions, model, format, duration, and total estimate before creating the video.

After confirmation, the Agent generated one playable 15-second MP4 and saved it with the product and demonstrator references in Project Assets. The completed output kept the opening, dropper, facial application, radiance, packshot, and closing brand beats in the planned order.

Completed Agent project with the generated 15-second marketing video in Project Assets — The Agent reports completion and adds the generated video to the project.

Generated LUMIERE serum packshot near the end of the Agent video — Preview the finished video and check the product packshot, label, action sequence, and closing frame before export.

Before publishing a product video: replace demonstration assets with your own approved product materials and verified facts. Inspect packaging geometry, application actions, skin appearance, and every readable word. For exact brand or offer text, add a controlled text layer in the editor instead of relying on text rendered inside generated footage.