UniRoute
Scene Template · Text-to-VideoFixed Model: wan/2-5-text-to-video · Option: 720p:5

Earth Zoom Out AI Video Template (Wan 2.5 Text-to-Video)

Enter any city landmark below to assemble a 5-second street-to-space camera pullback prompt on Wan 2.5 (720p:5) with live gateway pricing.

Open Wan 2.5 Text-to-Video Model PageJump to Prompt & Input Presets

Build & Run Your Earth Zoom Out AI Prompt

Enter a landmark and city below or pick a preset example. The page shows the catalog price of wan/2-5-text-to-video at fixed option 720p:5 before sign-in, then your live USD quote and wallet balance before running. Review your options and price before generating.

Fixed Model: wan/2-5-text-to-video · Option: 720p:5Wan 2.5 Text-to-Video
Quick-load preset inputs:
59/240

Name a recognizable landmark and city so the continuous pullback transitions from street level to skyline, continent, and orbital Earth.

Assembled English Prompt517/2000 chars

One continuous unbroken camera shot starting at street level in the Eiffel Tower and Champ de Mars gardens in Paris, France, smoothly and rapidly pulling straight back and zooming out vertically through golden-hour clouds, revealing the surrounding city skyline, coastline, continent, and finally the glowing curvature of planet Earth suspended in deep starlit space. Single seamless zoom-out transition, bright clear daylight, crisp architectural and geographic details, no cuts, no black screen, no text, no labels.

Locked model: wan/2-5-text-to-video·Locked option: 720p:5 (720p resolution · 5-second clip (fixed option 720p:5))

The selected option’s price is shown below; generated results can be played and downloaded.

Catalog price: $0.375 USD · as of

Live Session Output
Ready

Play and download your result here after generating.

The selected option’s price is shown below; generated results can be played and downloaded.

Billing note: Funds are reserved up to your confirmed quote and charged only when settled; released holds return to your balance automatically.
Earth zoom out AI video template illustration: frames pulling back from the Eiffel Tower to the Paris skyline, Europe and planet Earth

This Earth zoom out AI video template builds a single-shot camera pullback prompt for Alibaba Wan 2.5 text-to-video (wan/2-5-text-to-video, fixed 720p:5 option). Enter one city landmark or geographic location, and the template wraps it in an unbroken vertical ascent sequence—from street level through golden-hour clouds to the city skyline, coastline, continent, and planet Earth in space.

Important scope note: this page generates a creative AI video synthesis from text descriptions, not a real satellite map, GIS orthophoto, or Google Earth Studio export. It is designed for creators who want a fast text-driven pullback prompt without manually chaining scale-transition keywords.

Landmark Pullback Prompt Presets (Text Examples)

Inspect how a single landmark input expands into a full multi-scale camera pullback prompt. Click a preset button in the generator above to populate the input field.

Note: These cards are structured prompt and input examples, not pre-rendered video outputs or verified camera trajectories. Generative text-to-video models synthesize artistic transitions that can vary by run.

Prompt Preset #1wan/2-5-text-to-video · 720p:5

Paris Eiffel Tower to Orbital Earth Pullback

Continuous vertical pullback starting at the Eiffel Tower and Champ de Mars, rising through golden-hour clouds to reveal Europe and Earth in space.

Preset illustration: the Eiffel Tower at golden hour beneath clouds and the curve of planet Earth
Illustration
Structured Input Fields
City Landmark or Geographic Location: the Eiffel Tower and Champ de Mars gardens in Paris, France
Assembled English Prompt
One continuous unbroken camera shot starting at street level in the Eiffel Tower and Champ de Mars gardens in Paris, France, smoothly and rapidly pulling straight back and zooming out vertically through golden-hour clouds, revealing the surrounding city skyline, coastline, continent, and finally the glowing curvature of planet Earth suspended in deep starlit space. Single seamless zoom-out transition, bright clear daylight, crisp architectural and geographic details, no cuts, no black screen, no text, no labels.
Text preset example (load into generator above)
Prompt Preset #2wan/2-5-text-to-video · 720p:5

Manhattan Skyline to Blue Marble Zoom Out

Unbroken camera ascent from Central Park and Midtown Manhattan towers through high-altitude clouds to the Atlantic coastline and planet Earth.

Preset illustration: the Manhattan skyline at golden hour under the Blue Marble Earth
Illustration
Structured Input Fields
City Landmark or Geographic Location: Central Park and the Midtown Manhattan skyline in New York City
Assembled English Prompt
One continuous unbroken camera shot starting at street level in Central Park and the Midtown Manhattan skyline in New York City, smoothly and rapidly pulling straight back and zooming out vertically through golden-hour clouds, revealing the surrounding city skyline, coastline, continent, and finally the glowing curvature of planet Earth suspended in deep starlit space. Single seamless zoom-out transition, bright clear daylight, crisp architectural and geographic details, no cuts, no black screen, no text, no labels.
Text preset example (load into generator above)

How to Prompt an Earth Zoom Out AI Shot in 3 Steps

This template structures the zoom-out prompt around intermediate altitude layers to guide the model from street level toward an orbital view (prompt intent only; actual visual continuity is not pre-verified and varies by run).

1. Enter a Recognizable Landmark & City

Provide both the street-level anchor and the broader city/region (for example, "the Eiffel Tower and Champ de Mars gardens in Paris, France") so the prompt includes cues for both near and mid altitudes.

2. Review the Multi-Layer Pullback Prompt

The template automatically chains the altitude sequence—street level, golden-hour clouds, skyline, coastline, continent, and orbital Earth curvature—and appends prompt requests for no cuts, no black screen, no text, no labels.

3. Review the Price and Run

The catalog price is available before sign-in; your account shows a live USD quote and wallet balance before you run. Submitting places a temporary hold that is released automatically if the upstream task fails.

Why Use a Structured Scale-Transition Template vs. a Blank Prompt?

Rather than submitting a bare phrase like "zoom out from Paris to Earth", this template expands your input with intermediate altitude cues (street -> clouds -> skyline -> coastline -> continent -> orbital curvature) and appends no cuts, no black screen, no text, no labels. These clauses describe the intended camera trajectory to Wan 2.5, though actual transition smoothness is not pre-verified and varies across runs.

Unlike photo-based effects (such as Higgsfield AI Earth zoom out workflows that start from an uploaded portrait or street photo), this template is a pure text-to-video prompt builder. It requires no image upload and focuses on architectural and geographic landmarks.

DimensionThis Scene TemplateBlank Model Prompt Box
Source input1 text field (Landmark + City name, no photo upload)Manual multi-sentence altitude choreography
Scale-transition cuesPre-wired street -> cloud layer -> skyline -> continent -> orbital Earth sequenceWritten manually from scratch each time
Visual natureCreative AI text-to-video synthesis (not satellite GIS data)Creative AI text-to-video synthesis (not satellite GIS data)
Model & option lockwan/2-5-text-to-video fixed at 720p:5 (5s)Manual selection on the model page

Fixed Model & Gateway Execution Details

Every task submitted from this template uses the standard UniRoute media gateway pipeline with live pricing and idempotent task recovery.

Runtime Model ID
wan/2-5-text-to-video (Alibaba Wan 2.5 Text-to-Video)
Fixed Option Key
720p:5 (720p resolution, 5-second duration)
Accepted Input Mode
Text prompt only (1 Location field -> <= 2000 char English prompt)
Pricing & Billing Basis
Catalog price before sign-in; live per-task quote and wallet balance in your account; hold is released if the provider fails

Scope & Usage Boundaries

  • Creative AI synthesis, not satellite imagery: The generated clip is an artistic AI video transition, not real geospatial imagery or Google Earth data.
  • Text-only input (no photo upload): Unlike image-to-video zoom-out filters (e.g., Higgsfield AI Earth zoom out photo effects), this template uses wan/2-5-text-to-video and does not accept uploaded photos.
  • Unverified single-shot consistency: Generative video models may still vary in how smoothly they bridge cloud and orbital layers across individual runs.
  • Paid execution only: Sign-in and USD wallet balance are required. If an upstream task fails, the held quote is released back to your balance.

Frequently Asked Questions

How does this differ from Higgsfield AI Earth zoom out or photo zoom effects?

Photo-based Earth zoom out tools require uploading a starting selfie or street photograph and running an image-to-video model. This UniRoute template uses wan/2-5-text-to-video in pure text-to-video mode: you enter a landmark and city name in text, and the model synthesizes the entire pullback scene without any photo upload.

Does this template use real satellite or Google Earth map data?

No. Wan 2.5 generates a creative cinematic representation of the landmark, skyline, coastline, and planet Earth based on learned visual patterns. It should not be used for geographic surveying or navigation.

Why does the prompt specify golden-hour clouds, coastline, and continent?

Listing intermediate altitude cues in the prompt is intended to guide the model through a gradual scale transition from street level to outer space rather than a single abrupt jump, though actual output behavior is not guaranteed.

Is the Earth zoom out AI template free?

No. UniRoute does not offer a free tier. Before sign-in, the page shows the catalog price for wan/2-5-text-to-video (720p:5); your account shows a live USD quote and available wallet balance before you run.

Am I charged if the upstream video task fails?

No. Creating a task places a temporary hold for the quoted USD amount. If the upstream provider fails to produce a video, the hold is automatically released (released state) and $0 is charged.

Can I run the same Earth zoom out prompt at 1080p or 10 seconds?

Yes. Copy the assembled prompt from the preview box on this page, then open the Wan 2.5 model page (/models/wan-2-5) where 720p:10, 1080p:5, and 1080p:10 options are available.

Related Models & Scene Templates

Wan 2.5 Text-to-Video Model Page

Access all four Wan 2.5 resolution and duration options (720p:5 to 1080p:10).

AI Cat Video Generator Template

Structured two-field Wan 2.5 prompt builder for indoor feline clips.

AI ASMR Video Generator Template

Macro tactile ASMR prompt template powered by Veo 3.1 (8s duration).