Skip to main content
Digen AI video and image workspaceDigen AI

Video workflow guide

How to Plan a Video with Digen AI

Prepare a prompt or a reference image, choose a video model and output settings, then inspect the result before downloading or refining the request.

Create an AI VideoRead the Steps

2 modes

Text-to-video and image-to-video

3 formats

Portrait, landscape, and square

Online

Preview results and check download access

Workflow

From Input to Review

Step 1

1 min

Choose text-to-video or image-to-video

Open the Digen AI workspace. Choose text-to-video for a written scene or image-to-video to guide the request with a reference image.

  • Use text-to-video when the scene starts as an idea.
  • Use image-to-video when a source image should guide the subject or composition.

Step 2

1-2 min

Write a clear motion prompt

Describe the subject, action, environment, camera movement, lighting, and mood. For example: 'A red sports car turns onto a wet city street at night, low tracking shot, neon reflections, cinematic lighting.'

  • Use action verbs such as turns, rises, walks, or zooms.
  • Describe one main camera movement instead of combining several conflicting moves.

Step 3

1 min

Add a reference image when needed

For image-to-video, upload a supported PNG, JPG, or WEBP image and describe the intended subject or camera movement. Keep the motion request consistent with the source image.

  • Use a sharp image with a clear subject.
  • Avoid asking the subject to change identity, shape, and setting at the same time.

Step 4

1 min

Choose the model, duration, and aspect ratio

Select an available video model and clip duration. Use 9:16 for TikTok, Instagram Reels, and YouTube Shorts; 16:9 for YouTube, websites, and presentations; or 1:1 for square placements.

  • Match the aspect ratio to the channel before generating.
  • Processing time and available duration depend on the selected model.

Step 5

Varies by model

Generate and review the video

Submit the request and follow its status. Inspect motion, composition, and framing. If needed, revise one prompt detail before trying again.

  • Change one variable at a time so you can see what improved.
  • Simplify the motion request if the scene becomes unstable.

Step 6

1 min

Download and publish the result

Review the completed video and check your plan's download access. Save an available result for a social clip, product scene, presentation, or further editing.

  • Save successful prompts for future campaign variations.
  • Check framing, brand details, and product claims before publishing.

Pro tips

Writing a Video Prompt

Describe the subject and motion first

A useful video prompt describes what is in the scene and what should change during the clip.

  • Start with the subject and one clear action.
  • Add the environment, lighting, mood, and visual style.
  • Use concrete movement such as walking, turning, drifting, or rising.

Control the camera and pacing

Camera language helps the video model understand how the viewer should move through the scene.

  • Try a close-up, wide shot, slow pan, tracking shot, or gentle zoom.
  • Use slow motion or fast pacing only when it supports the action.
  • Avoid stacking several camera moves into one short clip.

Write for the publishing format

Plan the composition around the channel where the finished AI video will appear.

  • Keep the subject centered for vertical TikTok, Reels, and Shorts.
  • Leave room for product details or captions in marketing videos.
  • Generate separate variations for different aspect ratios instead of cropping one result.

FAQ

Common questions before you start

How do I start a video request?

Choose text-to-video or image-to-video, add the prompt and any required reference, then review the model settings and cost. Sign in with sufficient credits to generate; downloads depend on your plan access.

What belongs in a text-to-video prompt?

Include the subject, action, environment, camera movement, lighting, mood, and style. Keep the main action clear and avoid combining too many scene changes in one short prompt.

How do I prepare an image-to-video request?

Upload a clear reference image, describe how the subject or camera should move, choose the model, duration, and aspect ratio, then generate the animated clip.

Which aspect ratio should I use for social media video?

Use 9:16 for TikTok, Instagram Reels, and YouTube Shorts, 16:9 for YouTube and websites, or 1:1 for square social placements.