Use Text, Images, Video, and Audio Together
Build a richer creative brief with more than a prompt. Add supported visual and audio references so H3 can use subjects, motion, scenes, sound, and other context when creating your video.
MiniMax H3 online
MiniMax H3 AI Video Generator
Create AI videos from text, images, video, and audio references with MiniMax H3. Guide motion, scenes, and sound in natural language.
Model overview
What Is MiniMax H3?
MiniMax H3 is a general-purpose multimodal generation model from MiniMax. It can understand context from text, images, video, and audio, then use those references to generate or edit audiovisual content.
H3 supports workflows such as text to video, image to video, reference-based generation, video editing, and V2V motion transfer. It can generate video with native stereo audio and supports output up to 15 seconds at up to 2K resolution.
For creators, marketers, e-commerce teams, and designers, this means you can describe a complete creative brief and use several types of reference media in the same workflow.
Features
One Model for Multimodal Video Creation
Describe Reference Relationships in Natural Language
Tell H3 what each input should do. Ask it to take camera movement from a reference video, a subject from an image, and sound direction from audio instead of treating every creative task as a separate workflow.
Generate Video with Native Stereo Audio
MiniMax H3 jointly generates audiovisual content and supports native stereo audio. This makes it useful for scenes that need voice, sound effects, music, or environmental sound as part of the generated result.
Create Up to 15-Second, 2K Video
MiniMax states that H3 can generate videos up to 15 seconds at up to 2K resolution. Higher-resolution output can be useful for product visuals, branded scenes, interface concepts, dynamic posters, and other detail-sensitive content.
Guide Motion with Video References
Use video as a motion reference for V2V motion transfer and other video-to-video workflows. Describe which movement, camera behavior, or action you want H3 to carry into your new scene.
Examples
MiniMax H3 Video Examples
Use cases
Built for Real MiniMax H3 Use Cases
Content Creators
Create short-form AI video for TikTok, Instagram Reels, YouTube Shorts, music visuals, personal projects, and storytelling concepts. Use text to video when you want to start from an idea. Add image, video, and audio references when you need more control over characters, motion, camera direction, or sound.
Marketing & Ad Teams
Build ad concepts, campaign hooks, product reveals, dynamic posters, social promos, and branded video drafts from a richer set of creative references. H3 can combine visual, motion, and audio context, making it useful when your campaign brief contains more than a single text prompt.
E-commerce & Brands
Turn product images and creative references into product videos, launch concepts, Shopify assets, promotional clips, and branded social content. Use image to video for simple motion, or add video and audio references when you want to guide camera behavior, pacing, sound, or overall presentation.
Enterprise & Industry
Explore product design presentations, UI/UX motion, game interfaces, concept videos, and visual communication for more complex workflows. H3's multimodal context lets teams describe relationships between reference assets in natural language before generating an audiovisual draft.
How to use
How to Use MiniMax H3 Online
Use MiniMax H3 in your browser through the PWZ generator. No local GPU, model weights, or ComfyUI setup is required.
- 01
Open the MiniMax H3 Generator
Select the MiniMax H3 generation workflow available on PWZ.
- 02
Add Your Prompt and References
Describe the scene you want. Upload supported image, video, or audio references that should guide the subject, camera, movement, style, or sound.
- 03
Explain What Each Reference Should Control
Write the relationship directly in your prompt. For example, specify which image defines the subject, which video guides movement, and which audio provides the sound direction.
- 04
Generate and Refine
Choose the available output settings, then generate your video. Review the result and adjust your prompt, references, or settings when you want another version.
Know the Cost Before You Generate
Review the required PWZ credit cost before generation so you can decide whether to continue.
Capabilities
MiniMax H3 Output and Model Capabilities
CapabilityMiniMax H3
Multimodal contextText, images, video, and audio
Text to videoSupported
Image / keyframe to videoSupported
Multimodal reference generationSupported
Video editing / reference workflowsSupported
V2V motion transferSupported
Audio outputNative stereo audio
Maximum video durationUp to 15 seconds
Maximum resolutionUp to 2K
Commercial scenarios highlighted by MiniMaxAdvertising, branding, e-commerce, product design, UI/UX, gaming
These are MiniMax H3 model capabilities. Features available through PWZ depend on the API or inference workflow implemented on this page.
Pricing
MiniMax H3 Pricing
Start free and buy PWZ credit packs when you need more MiniMax H3 generations. Credits never expire and include watermark-free downloads and commercial usage rights.
600 credits · $0.062 per credit
Perfect for trying out PWZ — get started with a credit balance for your first generations.
- 600 credits
- High-resolution output support
- Watermark-free downloads
- Commercial usage rights included
- Credits never expire
2,000 credits · $0.052 per credit
For creators generating regularly — better per-credit value for daily PWZ workflows.
- 2,000 credits
- Watermark-free downloads
- Commercial usage rights included
- Credits never expire
- 3.6× more credits than Starter
4,000 credits · $0.045 per credit
Best value for teams and content pipelines — more generations at the lowest per-credit cost.
- 4,000 credits
- Watermark-free downloads
- Commercial usage rights included
- Credits never expire
- 6.9× more credits than Starter
- Priority processing speed
FAQ
MiniMax H3 FAQ
What is MiniMax H3?
MiniMax H3 is a general-purpose multimodal generation model from MiniMax. It understands text, images, video, and audio context and can generate audiovisual content with native stereo audio.
Can I use MiniMax H3 online on PWZ?
Yes. PWZ provides third-party browser-based access to supported MiniMax H3 video generation workflows without requiring you to deploy the model locally.
Can MiniMax H3 generate video from text?
Yes. H3 supports text-to-video generation. You can describe the scene, motion, camera behavior, and other creative direction in your prompt.
Can I use images, videos, and audio as references?
Yes. H3 supports multimodal reference workflows using images, video, and audio. The exact number, size, and file formats accepted by PWZ depend on the implemented generation mode.
Does MiniMax H3 generate audio with video?
Yes. H3 supports native stereo audio as part of its audiovisual generation design rather than requiring every workflow to add audio only after video generation.
What resolution and video length does MiniMax H3 support?
MiniMax states that H3 supports video up to 2K resolution and up to 15 seconds. Resolution and duration options available through PWZ may vary by generation mode.
What makes MiniMax H3 different from a basic text-to-video workflow?
H3 is designed around multimodal context and task generalization. It can use relationships between text, images, video, and audio instead of relying only on a single text prompt.
Ready to create
Create Your First MiniMax H3 Video
Bring your prompt, images, video, and audio references into one online workflow. Describe how they should work together, then generate your next AI video with MiniMax H3 on PWZ.
Try MiniMax H3 Free