US

Hailuo - AI Video Generator

Explore API
MODEL

Choose an AI palette for artistic style.

Model Image
Hailuo 2.3 Fast

Image to Video with Camera Control

Nothing here yet.

Your assets will appear once published.Go To Playground

Hailuo — AI Video Generator

Camera Control Commands (Hailuo 2.3 Fast)

Hailuo 2.3 Fast uses square bracket syntax for camera direction. Place these commands at the start of your prompt before describing the scene content. Each command maps to a specific camera movement. You can combine up to 3 camera directives per prompt.

[Truck]

Camera moves horizontally left or right while maintaining its distance from the subject. Reveals elements on the sides of the frame while keeping the focal distance constant.

[Pan]

Camera rotates horizontally around a fixed position. The view sweeps left or right from the camera's stationary point. Good for revealing off-screen elements.

[Push/Pull]

Camera moves forward (push) or backward (pull) toward or away from the subject. Creates depth and changes perspective as the camera approaches or retreats.

[Pedestal]

Camera moves vertically up or down while maintaining its distance from the subject. Reveals vertical elements above or below without tilting the camera.

[Tilt]

Camera angles upward or downward while position remains fixed. Creates a sense of scale on tall structures, skies, or ground-level details from above.

[Zoom]

Optical zoom effect without camera movement. Tightens or widens the frame around the subject. The perspective does not change, only the framing magnification.

[Shake]

Camera vibration or tremor effect. Adds urgency, tension, or handheld realism. Works best for action sequences or cinematic emphasis.

[Tracking]

Camera follows a subject as it moves through the scene. The camera maintains a consistent distance while moving along with the subject.

[Static]

Camera remains completely stationary. No movement occurs. Use for establishing shots or scenes where camera motion would be distracting.

Note: Combine up to 3 camera directives per prompt using square brackets. Example: [Truck left, Pan right, Zoom in]. Combining more than 3 directives often produces conflicting motion.

Model Comparison

Model Input Type Base Image Required Prompt Limit Camera Control
Hailuo 2.3 Fast Image + Video Instructions Yes (JPG/PNG, ≤20 MB) 2000 characters Up to 3 bracket directives: [Truck, Pan, Push/Pull, Pedestal, Tilt, Zoom, Shake, Tracking, Static]
Hailuo 2.3 Pro Text only No 2000 characters No bracket directives available. Text-only video generation with cinematic realism focus.

Use Cases

Honest Limitations

LIMIT: Camera commands are approximate, not precise. [Push/Pull] does not guarantee a specific distance or speed. Results vary between generations.
LIMIT: No audio generation. All output is silent video. Sound must be added externally in post-production.
LIMIT: Complex scenes with multiple moving subjects often produce artifacts — subjects may merge, clip through each other, or deform during motion.
LIMIT: Text and fine detail degrade during video generation. Do not expect text readability in output.
LIMIT: No frame-by-frame editing. You get one continuous clip per generation. If a single frame has an artifact, regenerate entirely.
LIMIT: Combining more than 3 bracket directives per prompt produces unreliable results. The camera may ignore directives or produce conflicting motion.
LIMIT: Hailuo does not support multi-shot sequences. Each generation is a single continuous shot. Multi-shot videos must be assembled from separate generations.
LIMIT: Hailuo 2.3 Fast image uploads are limited to 20 MB. Larger files must be compressed before upload.

Frequently Asked Questions

What is Hailuo?
Hailuo is an AI video generator developed by MiniMax that specializes in image-to-video generation with precise camera control (Hailuo 2.3 Fast) and text-to-video generation with cinematic realism (Hailuo 2.3 Pro). Pixazo provides access through its platform.
What camera directives does Hailuo 2.3 Fast support?
Hailuo 2.3 Fast supports 9 camera directives: Truck, Pan, Push/Pull, Pedestal, Tilt, Zoom, Shake, Tracking, and Static. Use square bracket notation like [Truck left, Pan right, Zoom in]. You can combine up to 3 directives per prompt.
What is the difference between Hailuo 2.3 Fast and Hailuo 2.3 Pro?
Hailuo 2.3 Fast requires a base image and supports camera control directives for image-to-video generation. Hailuo 2.3 Pro is text-only and focuses on cinematic realism without image input or camera bracket controls. Choose Fast for image animation with camera movement, Pro for pure text-to-video cinematic content.
What is the file size limit for Hailuo 2.3 Fast uploads?
Base images for Hailuo 2.3 Fast must be 20 MB or smaller. Supported formats are JPG and PNG. Larger files must be compressed before upload.
How many camera directives can I combine in one prompt?
You can combine up to 3 camera directives per prompt. Example: [Truck left, Pan right, Zoom in]. Combining more than 3 directives often produces unreliable results with conflicting or ignored motion.
What is the character limit for prompts?
Both Hailuo 2.3 Fast and Hailuo 2.3 Pro support prompts up to 2000 characters. This includes camera directives in brackets and all scene description text.
Who developed Hailuo?
Hailuo was developed by MiniMax, an AI research company. Pixazo provides access to Hailuo through its playground platform and does not own, train, or modify the underlying model.
Does Hailuo 2.3 Pro support image upload?
No. Hailuo 2.3 Pro is text-only and does not accept image uploads. It generates videos from text prompts only. For image-to-video generation with camera control, use Hailuo 2.3 Fast.
Can I generate multi-shot sequences with Hailuo?
No. Hailuo generates single continuous shots only. Each generation produces one clip with one camera movement. For multi-shot videos, generate separate clips with different directives and edit them together in post-production.
With Pixazo’s platform we deliver enterprise-class security and compliance to you and your customers through every interaction.