Skip to main content

How to use Playground

A step-by-step guide to creating AI images and videos in Playground, from adding references and prompts to picking models and generating your final output.

Written by Vita

Want to create something completely your own? Playground gives you a flexible space to create your own images and videos with AI. You can start with a prompt, bring in one or more reference images, or put together an image first and turn it into a video.

Here’s how to do it:

  1. Log in to your Zeely account: https://app.zeely.ai

  2. From the left menu, go to Image ads, then pick out Playground from the tiles at the top.

  3. Choose what you want to create.

    Playground has two sections: Image and Video.

    • To create an image, stay on the Image tab.

    • To create a video straight away, switch over to the Video tab and build it from a starting image.

    • You can also put together an image first, then switch to Video and turn it into a video from there.

  4. Add your image references.

    On the Image tab, click the add image icon to open the reference picker. From here you can either upload a new file or pull one in from your existing gallery.

    • To upload: drag and drop a file into the upload area, or click it to browse your device. Zeely accepts WEBP, PNG, JPEG, JPG, and HEIC files up to 10 MB each. If you upload your own image, you could be asked to accept Zeely’s Terms of Use.

    • To reuse an existing image: browse the left-hand panel (Uploads, Projects, Assets) or pull one from your gallery under Image ads, Video ads, Viral videos, or Avatars.

    • You can select several images at once. As you tap through them, each one gets numbered in the order you picked it, which sets how they'll show up as references.

    ​Once you're done picking, click Continue with [number] images to bring them into your prompt.

  5. Tell AI how to use your references.

    Each image you add shows up above the prompt box as a small tag, like @img1 or @img2. You can call these out directly in your prompt to tell the AI exactly what each one should be used for.

    For example: "Use @img1 as the main subject, holding @img2, in the editorial style of the scene."


    That lets you line up one reference for a person's appearance, another for a product, and another for the overall style or composition.

    Tagging is optional. If you leave your prompt without any @ tags, all your selected images still get sent to the model, and it works out on its own how to use them.

  6. Describe your idea. Type your prompt into the text box. You can start from scratch or walk through how you want your references to appear in the scene, covering things like what the subject should be doing, where it's set, the mood, the lighting, and the overall style.

  7. Open the AI model dropdown and pick a model. The settings available underneath change depending on which one you choose:

    • Nano Banana Pro: 1K, 2K, or 4K resolution

    • GPT Image: 1K, 2K, or 4K resolution, plus Low, Medium, or High quality

    Below the model picker you'll also find:

    • Aspect Ratio: sets the shape of your image (square, portrait, landscape, and so on). Some ratios are flagged for specific uses, like 9:16 for Stories and Reels, or 4:5 and 1:1 for feed posts and ads. Default is 9:16.

    • Output count: how many versions to generate in one go, from 1 up to 4.

    If you switch models partway through, your prompt and references carry over. Should the new model not support a setting or reference you've already picked, Zeely will flag it and suggest alternatives instead.

  8. Once everything's set, click Generate. Each generation draws down a credit cost, shown right on the button.

    Nano Banana models usually take about a minute to generate; GPT Image takes closer to three minutes.

    When your images come through, look them over in the Projects panel on the right and pick out the one you want to keep working with.

  9. To bring an image to life, open it and click Create video to jump over to the Video tab with that image already set as your starting frame.


    You can also skip image generation altogether and head straight to the Video tab, then add a starting image from there.

    If you want your video to land on a specific final shot, click End frame and add an image for that too.

  10. Describe your video. Type your scene into the prompt box.

    You can lay out a single scene, or chain several together by labeling them Scene 1, Scene 2, Scene 3, and describing what happens in each.

    Go into as much detail as you need, covering characters, actions, setting, camera movement, mood, or transitions between shots.

  11. Open the AI model dropdown to compare your options:

    • Kling 3.0 (recommended) — built for advanced realism with cinematic camera movement, 1080p, 10s clips.

    • Seedance 2.0 — reference-guided motion with native audio support, 1080p, 10s clips.

    • Seedance 2.5 — faster generation with built-in audio support, up to 720p, 10s clips.

    Each option shows its resolution, default length, and credit cost right in the dropdown, so you can weigh them against each other before picking one.

    Once you've chosen a model, fine-tune the settings underneath:

    • Aspect ratio (or Auto, depending on the model)

    • Resolution

    • Duration — drag the slider anywhere from 4 to 30 seconds, depending on what the model supports

    • Sound — toggle on or off

    When you're happy with your setup, click Generate.

  12. Review, regenerate, or download your video. Once a video finishes processing, click on it to open the preview.

    From here you can:

    • Play it back using the controls under the video and check the timestamp and total length.

    • Look over the details on the right, including the filename, when it was created, the full prompt you used (with any tagged references, like a start frame, called out inline), and the settings you generated it with (model, sound, duration, resolution).

    • Copy the prompt by clicking the copy icon next to "Prompt," which comes in handy if you want to reuse or tweak it for another generation.

    • Regenerate the video with the same prompt and settings if you want another take. This draws down credits again, at the same cost as the original generation.

    • Download the file straight to your device.

    • Favorite or delete it using the star icon and the three-dot menu at the top of the panel.

You can use Playground whenever you want to experiment with a specific visual idea, combine several references, or turn a generated image into a video. Describe your idea and watch it turn into a creative piece ✦

If something doesn’t look right or you need a hand, reach out to our Support team through the in-app chat or email us at [email protected].

Did this answer your question?