MiniMax H3 · Hailuo 3 · online
MiniMax H3 AI Video Generator
Describe the shot: subject, action, camera, light and sound.
Credits go back to your balance if a render fails.
Turn a prompt, a picture or a few reference clips into a video with matching sound. Up to 15 seconds, native 2K, no editing.
- Text · image · reference
- Sound included
- Up to 15 s
- No watermark
- Credits back if it fails
Made with MiniMax H3
Real renders from this site, with the exact prompt. Use one as a starting point.
“A street dancer spins on a rooftop at dusk, the camera circles her, city lights flicker on below, a punchy hip-hop beat”
Text · MiniMax H3 · 2K · 5s
Use this prompt“A golden retriever in a tiny yellow raincoat splashes through puddles on a city street, slow motion, water droplets sparkle, cheerful ukulele music”
Text · MiniMax H3 Max · 1080P · 6s
Use this prompt“Macro shot of coffee being poured into a glass cup of iced milk, swirling clouds of cream, warm morning light, soft café ambience”
Text · MiniMax H3 · 768P · 5s
Use this prompt“A tiny orange kitten chases a soap bubble across a sunlit wooden floor, slow motion, playful piano music”
Text · MiniMax H3 Max · 480P · 5s
Use this prompt“The paper boat starts to spin slowly as the rain gets heavier, the camera pulls back to reveal the whole street”
Image · MiniMax H3 Max · 480P · 5s
Use this prompt“In the style and mood of the reference video, a yellow paper boat sails across a puddle in a sunny park, children laughing”
Reference · MiniMax H3 Max · 480P · 5s
Use this prompt
What is MiniMax H3?
MiniMax H3 (also called Hailuo 3) is MiniMax's general-purpose multimodal video model. It reads text, images, video and audio together and generates video with native stereo sound, up to 15 seconds at 2K.
This site runs MiniMax H3 and its faster variant, H3 Max, in the browser: write a prompt, add a first or last frame, or give it reference images, clips and audio, then download the MP4.
MiniMax H3 AI is an independent service and is not affiliated with or endorsed by MiniMax. MiniMax, Hailuo and MiniMax H3 are trademarks of their owners.
MiniMax H3 at a glance
Checked against the sources linked on each line.
- Made by
- MiniMax, the lab behind the Hailuo video models. MiniMax
- Released
- July 31, 2026. MarkTechPost
- Inputs
- Text, images, video and audio, understood as one context. MiniMax
- Output
- Up to 15 seconds at 2K, with native multi-shot scenes. MiniMax
- Sound
- Native stereo audio; voice, effects and music generated together. MiniMax
- Official app
- Also available in MiniMax's own Hailuo AI app. Hailuo AI
What you can make
One model for prompts, pictures and references, with the sound made in the same pass.
Native 2K
H3 renders 2K directly instead of upscaling, so faces, fabric and small text stay sharp.
Sound with the picture
Dialogue, effects and ambience are generated with the video and timed to the action. No separate audio step.
Up to 15 seconds
Long enough for several shots in one clip; pick any length from 4 to 15 seconds.
First and last frame
Start from your image, end on another, and describe what happens in between.
Reference images, clips and audio
Up to 9 images, 3 clips and 3 audio files guide the character, style, motion or voice.
Fast drafts with H3 Max
The fast variant renders 480P–1080P in about a minute, for trying ideas cheaply.
How it works
Three steps, a couple of minutes.
- 01
Describe the shot
Say what happens: subject, action, camera, light and sound. Or start from an example.
- 02
Add pictures or references
Optional: a first or last frame, or reference images, clips and audio to follow.
- 03
Generate and download
Pick the model, resolution and length, see the price, and download the MP4 when it's ready.
H3 or H3 Max?
Same model family: Max is faster and cheaper, H3 is sharper.
| MiniMax H3 Max | MiniMax H3 | |
|---|---|---|
| Resolution | 480P, 768P, 1080P | 768P, native 2K |
| Length | 5–15 s | 4–15 s |
| Inputs | Text, frames, references | Text, frames, references |
| Typical wait | About a minute | A few minutes |
| Price | From 6 credits/s | From 9 credits/s |
Pay per second
300 credits for $9.99. MiniMax H3 Max at 480P is 6 credits a second; a 5-second clip is 30 credits.
Starter
$9.99
300 credits · one-time
- ≈ 50 s of MiniMax H3 Max at 480P
- ≈ 21 s of MiniMax H3 at 2K
- $1.00 per 5-second clip
Creator
Most popular$29.99
1000 credits · one-time
- ≈ 166 s of MiniMax H3 Max at 480P
- ≈ 71 s of MiniMax H3 at 2K
- $0.90 per 5-second clip
Studio
Save 16%$69.99
2500 credits · one-time
- ≈ 416 s of MiniMax H3 Max at 480P
- ≈ 178 s of MiniMax H3 at 2K
- $0.84 per 5-second clip
MiniMax H3 Max
5–15 sMiniMax's fast variant. Best for drafts and social clips; usually ready in under a minute.
| Resolution | Per second | 5 s video |
|---|---|---|
| 480P | 6 credits | $1.00 |
| 768P | 9 credits | $1.50 |
| 1080P | 19 credits | $3.16 |
Reference video is billed per second like output. The first 2 reference images are included, then 9 credits each.
MiniMax H3
4–15 sThe flagship model with native 2K output, for the sharpest detail and small text.
| Resolution | Per second | 5 s video |
|---|---|---|
| 768P | 9 credits | $1.50 |
| 2K | 14 credits | $2.33 |
Reference video is billed per second like output. The first 5 reference images are included, then 4 credits each.
One-time packs · credits never expire · back automatically if a render fails
MiniMax H3 questions
MiniMax H3, also called Hailuo 3, is MiniMax's multimodal video model, released on July 31, 2026. It generates video with native stereo sound from text, images, video and audio, up to 15 seconds at 2K.
No. MiniMax H3 AI is an independent site that runs MiniMax H3 and H3 Max. MiniMax's own app is Hailuo AI.
Every video uses credits, because each render costs real compute. The smallest pack is 300 credits for $9.99, and a 5-second MiniMax H3 Max clip is 30 credits. Failed renders cost nothing.
You pay per second: from 6 credits a second (MiniMax H3 Max, 480P) up to 14 a second (MiniMax H3, 2K). The form shows the exact total before you generate.
H3 Max is the fast variant: 480P to 1080P, 5 to 15 seconds, usually ready in about a minute. H3 is the flagship with native 2K and 4 to 15 seconds. Both take text, frames and references.
Up to 15 seconds. H3 starts at 4 seconds and H3 Max at 5.
Yes. H3 generates voice, effects and ambience together with the picture, timed to the action. Describe the sound you want in the prompt.
Yes. Choose Image, add a first frame (and optionally a last frame), and describe what happens. The video takes the shape of your first image.
Files the model follows for a character, style, motion or voice without using them as frames: up to 9 images, 3 clips and 3 audio files, clips 2 to 15 seconds and 15 seconds in total. Audio needs at least one image or clip.
Your credits go back to your balance automatically. Requests the provider rejects, for example for content, are not charged either.
Yes, the videos you make are yours to use, as long as your prompt and files don't infringe anyone's rights. See the terms for details.
They're used only to make your video and deleted 30 days later. Finished videos stay in your history until you delete your account.
Make your first MiniMax H3 video
Write a sentence, pick a length, and hear it come to life.
Start creating