What inputs does MiniMax H3 accept?

Use a text prompt, a start image with an optional end image, or a reference set containing images, videos and audio. Audio references must accompany at least one image or video reference.

What duration and resolution can I choose?

Choose any whole-second duration from 5 through 15 seconds and render at 768P, 2K or 4K at 24 FPS.

How many references can H3 use?

Reference mode accepts up to 9 images, 3 videos and 3 audio clips, with no more than 12 unique references in one request.

Does MiniMax H3 generate audio?

Yes. H3 produces native stereo audio with the video; there is no separate audio switch.

Can I generate with MiniMax H3 from this page?

Yes. MiniMax H3 is available in MaxVideoAI, and the exact price is shown before every render.

AI video · MiniMax

MiniMax H3

Direct the scene. Shape the sound.

MiniMax H3 creates 5–15-second videos from text, opening or ending frames, or mixed references, at resolutions from 480P to 4K. Direct identity, motion and native stereo sound together, or add your soundtrack in text and image modes.

Generate with MiniMax H3 View MiniMax H3 examples

MiniMax H3 wildfire lookout rescue sequence with Lio, red flares and native stereo dialogue — Real MiniMax H3 render
Lio leaves a wildfire lookout tower to guide stranded hikers along a red flare rescue trail

See MiniMax H3 Max for output up to 1080P Compare with Kling O3 Pro View pricing Prompt examples

Three creation workflows

Move between text-to-video, image-to-video and reference-to-video inside one H3 engine.

5–15 seconds at up to 4K

Choose any whole-second duration and render at 480P, 768P, 2K or 4K at 24 FPS.

Multimodal reference direction

Combine image, video and audio references to guide identity, performance, motion and sound.

Native stereo audio

Use a soundtrack in text or image mode, or audio references in reference mode. Describe dialogue, ambience and synchronized sound cues in the prompt.

MiniMax H3 pricing for your render

Choose workflow, duration, resolution and references, then review the exact price before rendering.

Generate with MiniMax H3

5s · 768P · from text

Live quote

audio incl.

10s · 2K · from an image

Live quote

15s · 4K · with references

$3.12

audio incl.

10s · 2K · 9 reference images

$2.11

audio incl.

The first five reference images are included; each additional reference image adds its displayed surcharge.

Character: [Appearance, wardrobe, expression and voice] Setting: [Location, time, weather and visual anchors] Action: [One clear physical action and reaction] Sound: [Dialogue, ambience and synchronized effects]

Lighthouse keeper rescue sequence

Text-to-video

Subject: Elara, a lighthouse keeper in a mustard raincoat • Action: Races up the tower, powers the lens and calls a lifeboat home
Camera: Handheld exterior tracking, stairwell close follow, lantern-room orbit and aerial pullback • Style: Storm-lit cinematic naturalism at blue hour
2K render: 15-second 2K multi-shot scene with native stereo dialogue

View full prompt

Create a 15-second cinematic 16:9 film with native multi-shot continuity and native stereo sound. Original fictional character: ELARA, a 34-year-old lighthouse keeper with wind-burned olive skin, dark wavy hair tied back, a weathered mustard raincoat, charcoal sweater, and practical sea boots. Keep her face, clothing, and proportions consistent in every shot. 0.0–3.5 seconds: wide exterior at blue hour, a stone lighthouse on a violent Atlantic cliff, rain sweeping sideways, Elara runs toward the tower carrying a small weathered signal case; handheld tracking camera, realistic wet fabric and foot contact. 3.5–7.5 seconds: interior spiral stairwell, medium close tracking shot as she climbs fast, breath visible, one hand gripping the rail, practical amber lamps, no cuts that break identity. 7.5–11.5 seconds: lantern room, she reaches the controls and throws the emergency lever; orbit from her determined face to the huge Fresnel lens as it powers on. At 10.5 seconds Elara says clearly, with natural lip sync: “Harbor light is alive. Bring them home.” 11.5–15.0 seconds: exterior aerial pullback as the rescue beam sweeps across black waves and reveals a distant lifeboat turning toward safety. Audio: native stereo only—storm wind outside, rain on glass, boots on iron stairs, strained breathing, lever clunk, electrical rise, deep lighthouse mechanism, distant surf, and Elara's exact spoken line. No music. No subtitles, captions, on-screen text, logos, watermarks, beauty-ad framing, or product pack shot. Photorealistic skin and motion, coherent hands, restrained cinematic contrast.

MiniMax H3