WaveSpeed.ai
DashboardExploreInspiration
Home/Explore/Hailuo Video Models/minimax/hailuo-02/t2v-standard

text-to-video

MiniMax Hailuo 02 T2V Standard | 768P Text-To-Video With Physics Scenes | WaveSpeedAI

minimax/hailuo-02/t2v-standard

Hailuo 02 is a text-to-video model on MiniMax, fine-tuned to output responsive 768P videos even for complex physics-driven scenes. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

The model automatically optimizes incoming prompts to enhance output quality. This also activates the safety checker, which ensures content safety by detecting and filtering potential risks.

Idle

Your request will cost $0.23 per run.

For $10 you can run this model approximately 43 times.

One more thing::

ExamplesView all

README

Hailuo 02 — Text-to-Video (T2V Standard)

minimax/hailuo-02/t2v-standard converts pure text into cinematic video with strong prompt adherence, believable physics, and smooth scene transitions. Built on MiniMax’s evolving framework, it balances quality, speed, and cost for everyday creative production.

Why it works

  • 768p output — crisp frames, ready for editing.
  • 6s / 10s clips — fast ideation or trailer-style beats.
  • Enhanced motion & physics — debris, cloth, water, camera shake look natural.
  • Cinematic continuity — fewer artifacts; smoother temporal flow.
  • Repeatable — low randomness for the same prompts.

Limits & Performance

  • Resolution: 768p
  • Durations: 6s or 10s
  • Runtime: ~30–90 s per clip (scene complexity & queue dependent)

Pricing

DurationCost per jobMax Resolution
6 s$0.23768p
10 s$0.56768p

How to Use

  1. Prompt — Describe scene, lighting, motion, and camera.
  2. Duration — Choose 6s or 10s.
  3. (Optional) Prompt expansion — Enable for smarter parsing & safety.
  4. Run — Review, tweak prompt/seed, and iterate.

Pro Tips

  • Write like a director: shot type, camera move, lighting, mood.
  • Use action verbs: dolly in, orbit, rack focus, particles swirl.
  • Start with 6s for quick iterations; assemble multiple clips for longer stories.
  • Add your own music/VO for complete narratives.