Turn a still scene or written brief into one focused shot. Choose 6 or 10 seconds at 768p, with no generated audio. Online generation is not available yet.
Start frame
Match first frame
768p
01Switch models without leaving the shared workspace.
02See the final credit cost before every generation.
03Find finished work in your results and history.
What this setup covers
Two clips, two practical ways to begin.
same-input resultA real Movelume result from the shared comparison brief. Use it to inspect this model beside the other outputs, then choose a model before preparing your own draft.Crimson Beam in the FogA 5.875-second text-to-video study approaches a dark lighthouse through predawn fog. A narrow crimson beam ignites late in the shot, sweeps once across mist and salt spray, then settles while the tower and basalt sea stack remain stable.
Crimson Beam in the FogA 5.875-second text-to-video study approaches a dark lighthouse through predawn fog. A narrow crimson beam ignites late in the shot, sweeps once across mist and salt spray, then settles while the tower and basalt sea stack remain stable.
Direct the six seconds
Start with motion the source frame can support.
This page pairs a lighthouse scene created from text with an image-to-video object study. Together they show how a clear written scene and a clear first frame lead to different kinds of Hailuo 2.3 motion.
Start with motion the source frame can support.
A short image-to-video prompt works best as a sequence of visible decisions. Name the primary action, give the camera one job, describe a restrained environmental response, and close with what must remain stable. The featured clip follows that pattern; the writing examples below are planning ideas, not guaranteed results. Write for one short shot. A clear order of motion is easier to inspect than simultaneous turns, zooms, weather changes, lighting changes, and object transformations.
Delay the reveal so the change has meaning.
The lighthouse scene begins in predawn fog with the lantern dark. The crimson beam appears late in the clip, sweeps once through mist and spray, and becomes the single event that changes the shot. A timed reveal gives a short scene a beginning, turn, and resolved image without adding a second location or subject. State when the change begins and what the viewer should notice afterward.
Give the camera one steady route through the atmosphere.
A restrained push moves from a wide aerial view toward the lantern room while wind-driven fog and faint waves provide supporting motion. The camera does not orbit, shake, or jump closer. One camera direction makes the lighthouse easier to follow as the fog changes around it. Use environmental motion to create depth without competing with the central reveal.
Protect the structure while weather stays active.
The architecture, sea stack, horizon palette, and scale remain consistent while mist, spray, waves, and the narrow beam move. The scene gains atmosphere without asking the building itself to transform. Separate fixed geometry from moving weather in the brief. This helps preserve a strong landmark silhouette while still giving the clip visible depth and progression.
same-input resultA real Movelume result from the shared comparison brief. Use it to inspect this model beside the other outputs, then choose a model before preparing your own draft.
Where Hailuo 2.3 fits
Plan Hailuo tests where continuity can be judged clearly.
The example clips cover focused object motion and a short atmospheric scene. The additional ideas show how the same one-shot approach can work for faces, products, and places without asking a 6- or 10-second clip to carry a long story.
Subtle portrait motion
A gaze change, breath, or slight expression makes identity continuity easy to inspect. Keep the camera restrained and compare the first and final frames closely.
Focused product reveal
Use light travel, shallow parallax, or a small slide to test whether silhouette, materials, and label placement survive motion without demanding hidden geometry.
Layered environmental motion
Wind, mist, water, or cloud movement can test depth and temporal coherence. One main camera path keeps the result measurable rather than merely busy.
What this setup covers
Two clips, two practical ways to begin.
Hailuo 2.3 currently offers 6- or 10-second clips at 768p. Each displayed clip keeps the settings used to create it. Use the examples to set a realistic direction, then choose the duration available for your own starting input before drafting the shot.
Input
The image should already contain the subject, composition, lighting direction, and room needed for the motion. It is the first frame, not a loose style reference.
Motion brief
Direct the subject first, then the camera and environment. Keep identity, product shape, typography, or architecture explicit when continuity matters.
Tested output
Choose 6 or 10 seconds in the active workflow. Keep the requested action small enough to begin, read, and settle inside that selected duration instead of treating ten seconds as permission for several scenes.
Availability
One example begins with an image and tests object motion; the other begins with text and builds an atmospheric scene. Use them to plan your shot, while expecting a new subject or prompt to produce a different result.
Before you choose Hailuo 2.3
What this page does not promise.
Use the two clips as starting points for a focused shot. They show one image-led and one text-led direction, while a new subject, camera move, duration, or visual style may still behave differently.
Online generation is not available yet
You can choose Hailuo 2.3 in the image-to-video or text-to-video workspace and prepare a prompt with the settings currently shown there. Drafting does not upload a file or spend credits.
Two examples show two starting directions
The playable clips show one image-led brief and one text-led brief that produced usable results. Compare their focused subject motion and environmental storytelling, but judge every new result on its own because quality and consistency can vary.
The current setup starts at 70 credits
The final amount can change with the selected settings and will be shown before generation begins. A starting credit amount does not guarantee queue time, delivery speed, or that every request can be completed.
Use images you have permission to animate
Use only images you created, licensed, or have permission to transform. Review the result for rights, private information, deceptive use, and visible AI disclosure before publishing it.
Prepare without submitting
Prepare one clear six-second shot.
Start from an image you are allowed to use, describe one main motion and one camera behavior, then decide what must remain stable before you judge the result.
01
Choose the continuity challenge
Decide whether the test should emphasize identity, product form, camera obedience, or layered scene motion. One main risk makes the output easier to judge.
02
Prepare a rights-cleared frame
Use an image you own or may lawfully transform. Record its source and intended use before any future upload, and avoid public figures or unlicensed characters for calibration.
03
Write the six-second direction
Keep one subject action, one camera behavior, and explicit stability constraints. The image-to-video workspace can hold this shot plan, but it does not convert the request into Hailuo.
04
Review the whole clip before you use it
Check geometry, labels, subject consistency, camera restraint, light continuity, and the final frame. Only keep a result that meets the needs of the shot you actually planned.
Plans for image and video
Pricing
Annual plans are billed upfront. The same plan credits refresh every 30 days and expire at the end of that window; unused subscription credits do not roll over.
Free
$0/ month
Explore the workspace and prepare a first request before choosing a paid plan.
Common questions about Hailuo 2.3 inputs, example clips, output settings, starting credits, and current availability.
01Can I generate with Hailuo 2.3 on Movelume now?+
Not yet. You can open Hailuo 2.3 in the image-to-video workspace, prepare the prompt and settings, and keep the draft in this browser without uploading anything or spending credits.
02Which Hailuo 2.3 settings is Movelume evaluating?+
Choose image to video with one first frame, or text to video with a written brief. Both offer 6 or 10 seconds at 768p and one output without generated audio. The image-led result follows the source frame; text-led shots use 16:9.
03How many credits will this Hailuo setup use?+
A 6-second shot costs 70 credits; a 10-second shot costs 140. Review the exact amount before generation. Preparing a draft uses no credits.
04Why is online Hailuo generation not available yet?+
Not yet. You can study the examples and prepare a shot without uploading a file or spending credits. The page will clearly show when Hailuo 2.3 becomes available to generate.
05Are the prompt examples guaranteed to work?+
No. They are writing patterns for focused Hailuo briefs, not guaranteed recipes. The two playable clips show their own image-led and text-led directions; the portrait and product ideas are planning examples that still need their own results and checks.
06What can I learn from the example clips?+
The two clips show successful Hailuo 2.3 directions for controlled subject motion and a short atmospheric scene. Use them to judge motion density, camera restraint, and scene stability, then review every new result on its own terms.
Prepare the shot first
Plan one clear Hailuo shot.
Choose image to video when a first frame must anchor the shot, or text to video when the scene can begin from a written brief. Select 6 or 10 seconds, keep one main action, and use the example clips to set a realistic motion target.
