MiniMax H3
Text & image to video · 2026-07
MiniMax's newest video model (Hailuo 3.0). Clips of up to 15 seconds with native audio and image, video and audio references. H3 Lite and H3 Highspeed are faster, lower-cost tiers.
AI video model · By MiniMax
Hailuo 02, Hailuo 2.3 and MiniMax H3 in your Qten account: lifelike motion and expressive characters, from text or images.
Free plan · No credit card · Updated 2026-11-18
MiniMax Hailuo
By MiniMax
7 versions of MiniMax Hailuo in one Qten account, from fast drafts to the newest release.
Hailuo AI is MiniMax's AI video generator. Hailuo 02 and 2.3 are known for realistic physics and expressive faces at up to 1080p, and MiniMax H3 (Hailuo 3.0) adds native audio and clips of up to 15 seconds. Qten offers seven versions, including the Fast, Lite and Highspeed tiers.
Hailuo is the video model line from the Chinese AI company MiniMax. Hailuo 02 (June 2025) made 1080p clips with strong physics, and Hailuo 2.3 (October 2025) improved motion, faces and stylized looks. MiniMax H3, also called Hailuo 3.0, launched in July 2026 with sound generated with the video, longer clips and reference inputs.
| Made by | MiniMax |
|---|---|
| Newest on Qten | MiniMax H3 |
| Versions on Qten | 7 |
| Max resolution | Up to 1080p |
| Max length | Up to 15 s (MiniMax H3) |
| Native audio | Yes (MiniMax H3) |
| Inputs | Text prompt · Start image · First & last frame · Reference images |
Limits are the maker's published figures, checked 2026-11-18. What you can use may depend on your plan.
7 versions of MiniMax Hailuo in one Qten account, from fast drafts to the newest release.
Text & image to video · 2026-07
MiniMax's newest video model (Hailuo 3.0). Clips of up to 15 seconds with native audio and image, video and audio references. H3 Lite and H3 Highspeed are faster, lower-cost tiers.
Text & image to video · 2025-10
Sharper motion and more expressive faces than Hailuo 02, at up to 1080p, in 6 or 10 second clips. Hailuo 2.3 Fast is the quicker, lower-cost option.
Full Hailuo 2.3 pageText & image to video · 2025-06
The June 2025 release that made Hailuo known for realistic physics: water, cloth and bodies that move believably.
| Version | What it makes | Tier | Released |
|---|---|---|---|
| Hailuo 02 | Text & image to video | Standard | 2025-06 |
| Hailuo 02 Fast | Text & image to video | Fast / lower cost | — |
| Hailuo 2.3 | Text & image to video | Standard | 2025-10 |
| Hailuo 2.3 Fast | Text & image to video | Fast / lower cost | — |
| MiniMax H3 | Text & image to video | Standard | 2026-07 |
| MiniMax H3 Lite | Text & image to video | Fast / lower cost | — |
| MiniMax H3 Highspeed | Text & image to video | Fast / lower cost | — |
Hailuo handles physics, sport and fast action with natural movement.
Faces and body language that carry emotion, in realistic or stylized looks.
MiniMax H3 generates dialogue, effects and ambience with the picture.
Hailuo 2.3 Fast, H3 Lite and H3 Highspeed for quick drafts and volume.
A gymnast performs a backflip on a balance beam in slow motion, chalk dust in the spotlight, crowd gasps, tracking camera, 16:9.Close-up of an old fisherman reading a letter on a pier, his eyes fill with tears, then he smiles, waves lapping, golden hour, 9:16.Anime-style girl with silver hair runs through a field of sunflowers, petals flying, wind in her hair, bright summer sky, 16:9.Believable physics for movement-heavy shots.
Emotional close-ups for stories and ads.
Anime and illustrated looks with smooth motion.
Fast tiers for many short clips.
The other video models in the same Qten account. Pick per scene, or let Qten choose.
| MiniMax Hailuo | Kling | Seedance | Veo | Wan | |
|---|---|---|---|---|---|
| Made by | MiniMax | Kuaishou | ByteDance | Google DeepMind | Alibaba |
| Newest on Qten | MiniMax H3 | Kling 4 | Seedance 2.5 | Veo 3.1 | Wan 3 |
| Versions on Qten | 7 | 6 | 7 | 2 | 10 |
| Max resolution | Up to 1080p | Up to 1080p | Up to 1080p | Up to 1080p | Up to 1080p |
| Max length | Up to 15 s (MiniMax H3) | Up to 30 s (Kling 4) | Up to 30 s (Seedance 2.5) | Up to 8 s | Up to 30 s (Wan 3) |
| Native audio | Yes (MiniMax H3) | Yes | Yes (Seedance 2.0 / 2.5) | Yes | Yes (Wan 2.5+) |
| Inputs | Text prompt · Start image · First & last frame · Reference images | Text prompt · Start image · First & last frame · Reference images | Text prompt · Start image · Reference images · Audio | Text prompt · Start image · First & last frame · Reference images | Text prompt · Start image · Reference images · Audio |
Limits are the maker's published figures, checked 2026-11-18. What you can use may depend on your plan.
Sign up on the Free plan. No credit card, and you get credits to start.
Write your prompt or script, then choose MiniMax Hailuo as the model, or let Qten pick the best model for each step.
Review the result, edit it in Qten, and post or schedule it to YouTube, TikTok, Instagram and more.
FAQ
Hailuo AI is the AI video model line from MiniMax. It creates video from text or images and is known for realistic physics and expressive characters.
Qten offers Hailuo 02, Hailuo 02 Fast, Hailuo 2.3, Hailuo 2.3 Fast, MiniMax H3, MiniMax H3 Lite and MiniMax H3 Highspeed.
Qten has a Free plan with 100 free credits and no credit card. Generating with Hailuo uses credits, depending on the version and settings. See the Pricing page for details.
MiniMax H3, also called Hailuo 3.0, is MiniMax's video model launched in July 2026. It adds native audio, reference inputs and clips of up to 15 seconds.
Hailuo is a strong pick for physical action and emotional close-ups. Kling makes longer, multi-shot clips. Both are in Qten, so you can choose per scene.
No. You use Hailuo from your Qten account with Qten credits, without a separate subscription or API key.
Model names and logos are trademarks of their owners. Qten is not affiliated with or endorsed by them.
MiniMax Hailuo and 130+ other AI models in one account. Start free.
Start creating free