Hedra
  • Enterprise
  • Pricing
  • Blog
  • Creators
Log inSign Up
Open Hedra
Your account
    Explore
  • Home
  • Enterprise
  • Pricing
  • Blog
  • Creators
    Log inSign Up
    Open Hedra

MiniMax Hailuo-02 Pro

All video models
Video modelMiniMax
OverviewImage → VideoText → Video

Overview

MiniMax Hailuo-02 Pro is a professional-grade video generation model developed by MiniMax that produces native 1080p cinematic content from text or image prompts. Known for its advanced physics simulation and temporal stability, the model accurately renders complex dynamics, fluid motion, and precise camera movements. Offering sharper textures and greater detail than MiniMax Hailuo-02 Standard, it is well-suited for commercial advertising, character animation, and short films requiring realistic, director-level control.

MiniMax Hailuo-02 Pro Image to Video

Image → Video — generates video.

Specifications

Input mode
Image → Video
Accepts
start frame
Aspect ratios
16:9
Durations
6s
Max duration
6s
Native audio
No
Pricing
16 credits / second — longer clips and higher resolutions cost more
Typical generation time
~3 min
Free tier
No

Image → Video examples

A 3D animated scene showing a fluffy white sheep and an upright orange goldfish walking side-by-side along a stone path toward a grand mosque at sunset. This first frame of a 1934x1080 video, generated using the MiniMax Hailuo-02 Pro model, captures the whimsical characters from behind in warm, golden lighting.Sheep and Goldfish at a Mosque — MiniMax Hailuo-02 ProA digital animation frame at 1080x1080 resolution showing cartoon sea creatures having a tea party underwater, generated using MiniMax Hailuo-02 Pro. A pink octopus pours tea at a blue table for a purple seahorse and an orange sea monster, surrounded by glowing yellow flora and coral.Underwater Sea Creature Tea Party — MiniMax Hailuo-02 ProA first frame of a 1922x1080 video showing a young woman with long brown hair speaking directly into a black microphone with a white logo. The background is a softly blurred room with bookshelves, warm lamps, and indoor plants. Generated using the MiniMax Hailuo-02 Pro model.MiniMax Hailuo-02 Pro: Podcast Host Speaking into MicrophoneA man and a woman wear over-ear headphones and sit behind professional microphones in a recording studio. Generated using MiniMax Hailuo-02 Pro at 1894x1024 resolution, this frame shows the male host looking thoughtful while the female host smiles warmly under soft studio lighting.Podcast Recording Session in Studio — MiniMax Hailuo-02 ProA video still shows a man and a woman hosting a podcast in a studio. The woman gestures animatedly while speaking into a microphone, and the man smiles in profile. They wear headphones at a wooden table. Generated using MiniMax Hailuo-02 Pro at 1894x1024 resolution.Podcast Hosts Discussing Science, by MiniMax Hailuo-02 ProA medium shot of a man in a light blue t-shirt eating a hot dog with a napkin in a sunny, green park with blurred people in the background. Generated with MiniMax Hailuo-02 Pro at 1894x1024 resolution, this video frame captures a casual, outdoor lifestyle scene.Man Eating a Hot Dog in the Park — MiniMax Hailuo-02 ProA video still at 1894x1024 resolution generated using MiniMax Hailuo-02 Pro. A man in a blue button-down shirt and a woman in a black top sit across a wooden table in a podcast studio, speaking into black microphones. Glass bottles and papers rest on the table against a soft-focus background.Podcast Hosts in Studio Discussion — MiniMax Hailuo-02 ProA still image from an image-to-video generation by MiniMax Hailuo-02 Pro at 1894x1024 resolution. It shows a male and female host in a podcast studio. The man on the left wears headphones and speaks into a microphone, while the woman on the right smiles and listens.Podcast Hosts in Studio — MiniMax Hailuo-02 Pro

MiniMax Hailuo-02 Pro Text to Video

Text → Video — generates video.

Specifications

Input mode
Text → Video
Aspect ratios
16:9
Durations
6s
Max duration
6s
Native audio
No
Pricing
16 credits / second — longer clips and higher resolutions cost more
Typical generation time
~4 min
Free tier
No

What is MiniMax Hailuo-02 Pro best used for?

Hailuo-02 Pro generates native 1080p video with realistic physics. MiniMax built its Noise-aware Compute Redistribution (NCR) architecture to handle complex kinetic motions, such as gymnastics, fluid dynamics, and rigid-body collisions. Because it accurately simulates gravity and momentum, it is well-suited for sports footage, physical comedy, and action sequences without relying on upscaling.

When was Hailuo-02 Pro released, and what is its lineage?

Developed by the Chinese AI company MiniMax, Hailuo-02 Pro was officially announced on June 18, 2025. It succeeded MiniMax's original Video-01 model, bringing significant efficiency improvements and native 1080p support. In late October 2025, MiniMax superseded it with the MiniMax Hailuo 2.3 Pro and MiniMax Hailuo 2.3 Standard models, which further refined character micro-expressions, dynamic lighting, and stylized rendering.

How can I get the best results with Hailuo-02 Pro?

To leverage its physics engine, write prompts that explicitly describe weight, momentum, and material interactions—such as a heavy object splashing into water or a gymnast shifting their balance. For image-to-video workflows, providing both a start and end frame allows you to precisely keyframe transitions. This technique works well for generating controlled hyperlapses, virtual try-ons, and product motion graphics.

Similar models

Grok VideoxAIKling 2.1 MasterKlingKling 2.5 TurboKlingKling 2.6 ProKlingMiniMax Hailuo 2.3 ProMiniMaxHedra OmniaHedra

Prompt tips

  • Use bracket syntax for cameras: Explicitly direct the camera using bracketed commands like [Pan left] or [Dolly in] to trigger the model's trained cinematic movements.
  • Anchor with start and end images: For complex transformations, upload both a starting and ending image to lock the composition and guide the AI's transition across the scene.
  • Structure prompts for physics: Because the model relies on a strong physics engine, explicitly describe momentum, weight shifts, and material properties (e.g., "heavy landing," "fluid splash") to maximize kinetic realism.
  • Focus on facial expressions: The model excels at micro-expressions; specify the exact emotional shifts or subtle facial movements in your prompt rather than just describing the broader action.
Illustration of a laptop with the Hedra spark logo in front of a city skyline at sunset

What Will You Create?

Sign up for free

Product

StudioCommunityFeedbackUse CasesModels

Legal

Privacy PolicyTerms of useAcceptable useCookie PolicyBiometric data policy

Company

AboutTeamChangelogCareersCreatorsSupportAlternatives
LinkedinInstagramDiscord
support@hedra.comHedra 2026 — All rights reserved