MiniMax H3 API
MiniMax H3 API for cinematic text-to-video, image-to-video, and reference-to-video generation with strong motion quality, subject consistency, and scene coherence. Build social videos, ads, product showcases, character scenes, and scalable AI video workflows through WaveSpeedAI.
Generate cinematic AI videos from text prompts, still images, or visual references. MiniMax H3 gives creators and developers three focused API workflows for fast, flexible video production, from first concepts and social content to advertising, product storytelling, and character-led scenes.
About the MiniMax H3 API
What MiniMax H3 does, how it fits in the MiniMax model lineup, and why teams reach for it.
MiniMax H3 is a video generation model from MiniMax, available through the WaveSpeedAI REST API. MiniMax H3 API for cinematic text-to-video, image-to-video, and reference-to-video generation with strong motion quality, subject consistency, and scene coherence. Build social videos, ads, product showcases, character scenes, and scalable AI video workflows through WaveSpeedAI.
Generate cinematic AI videos from text prompts, still images, or visual references. MiniMax H3 gives creators and developers three focused API workflows for fast, flexible video production, from first concepts and social content to advertising, product storytelling, and character-led scenes.
Run MiniMax H3 through the same API key, billing account, and rate-limit envelope you use for the other 1,000+ AI models on WaveSpeedAI. No separate vendor setup, no per-provider SDKs, no per-vendor rate-limit envelopes — one integration covers everything from text-to-image and text-to-video through audio synthesis, 3D generation, upscaling, and editing.
MiniMax H3 API capabilities and release status
The model-specific details developers search for before choosing an API: availability, expected output length, reference support, and the current live fallback.
Text to video
Prompt-driven generation
Use minimax/h3/text-to-video to turn a written creative brief into a cinematic video with coherent motion, composition, and scene progression.
Image to video
Animate a still image
Use minimax/h3/image-to-video to bring a source image to life while carrying its subject, framing, visual identity, and art direction into motion.
Reference to video
Guide identity and style
Use minimax/h3/reference-to-video when visual references should guide the subject, character identity, style, or scene language of the generated video.
API integration
One WaveSpeedAI workflow
Submit H3 predictions with your WaveSpeedAI API key, track each request through the standard prediction lifecycle, and retrieve generated videos from the result response.
How to use the MiniMax H3 API
Four steps from signup to a finished generation. Full Python, Node.js, and cURL examples are in the API section below.
- 1
Get an API key
Sign up for a WaveSpeedAI account and copy your API key from the dashboard. New accounts come with free starter credits — enough to run the playground a few dozen times before billing kicks in.
- 2
Submit a prediction
POST your input as JSON to https://api.wavespeed.ai/api/v3/minimax/h3/text-to-video. The endpoint returns a prediction id immediately — generations are async so you don't hold an open connection during inference.
- 3
Poll for completion
GET https://api.wavespeed.ai/api/v3/predictions/{request_id}/result. Start around every 2 seconds, then increase toward 5-10 seconds for long-running tasks to reduce unnecessary requests. Stop on completed, failed, cancelled, or timeout.
- 4
Read the output URL
Once status is"completed", read the URL from data.outputs[0]. The URL points to your generated media on the WaveSpeedAI CDN — image, video, audio, or 3D file depending on the MiniMax H3 variant you called.
What you can build with MiniMax H3
Common workflows developers and creators use the MiniMax H3 API for.
Text-to-video for original concepts
Turn a prompt into an original cinematic scene with minimax/h3/text-to-video. Use it for concept visualization, storytelling, campaign ideas, and shots that do not need to begin from existing artwork.
Image-to-video for product showcases
Animate product photography, campaign stills, key art, or character images while preserving the visual direction of the source. Useful for product reveals, social ads, and motion-first landing content.
Reference-to-video for consistent subjects
Use visual references to guide character identity, subject appearance, scene design, or style. Reference-to-video is the H3 workflow for recurring characters, branded visuals, and connected creative sets.
Character scenes and visual storytelling
Create character-led scenes, story beats, mood pieces, and previsualization with an endpoint selected for the source material: text for new scenes, image for animation, or references for stronger continuity.
Scalable AI video generation pipelines
Integrate MiniMax H3 into applications and automated creative workflows through WaveSpeedAI. Use one API key and a consistent prediction lifecycle across all three H3 generation modes.
Tips for prompting MiniMax H3
Practical advice for getting better outputs from MiniMax H3 — drawn from the patterns that work across video models in production pipelines.
Choose the endpoint by source material
Use text-to-video for an original prompt, image-to-video to animate one source image, and reference-to-video when visual references should guide identity, subject appearance, scene design, or style.
Describe motion as a sequence
Write the starting state, action, environmental response, camera movement, and ending state in chronological order. A clear progression gives H3 a stronger motion plan than a list of disconnected visual adjectives.
Use image-to-video when composition is already approved
If the subject, product, framing, or art direction is locked in a still image, start with image-to-video instead of rebuilding the scene from text. Focus the prompt on movement and camera behavior.
Give references one clear job
For reference-to-video, state what the references should control: character identity, product appearance, visual style, environment, or composition. Avoid asking the same reference to define several conflicting aspects of the shot.
Keep one primary action per short scene
A focused action is easier to render coherently than a full sequence of unrelated events. Generate separate shots for separate story beats, then assemble them in editing when the concept needs more narrative scope.
MiniMax H3 vs alternatives
When to pick MiniMax H3 over similar models on WaveSpeedAI.
MiniMax H3 vs Hailuo 2.3
Hailuo 2.3 offers tiered text-to-video and image-to-video endpoints across Standard, Pro, Fast, and Fast Pro variants. MiniMax H3 uses a focused three-endpoint suite and adds reference-to-video as a first-class workflow for visual guidance and subject consistency.
MiniMax H3 vs Seedance 2.0
Seedance 2.0 provides a broad production family with text-to-video, image-to-video, video-edit, video-extend, native audio, and multiple performance tiers. MiniMax H3 is a simpler choice when the workflow centers on generating from text, a source image, or visual references.
MiniMax H3 vs Kling 3.0
Kling 3.0 emphasizes tiered output quality and dedicated motion-control endpoints. MiniMax H3 organizes its API around source type, including a dedicated reference-to-video route for identity, subject, and style-guided generation.
MiniMax H3 API — Frequently asked questions
Pricing, license, integration — common questions about running MiniMax H3 on WaveSpeedAI.
What is the MiniMax H3 API?
The MiniMax H3 API is a three-model AI video generation suite on WaveSpeedAI. It supports text-to-video for prompt-driven scenes, image-to-video for animating still images, and reference-to-video for generation guided by visual references.
Which MiniMax H3 API endpoint should I use?
Use minimax/h3/text-to-video when starting from a prompt, minimax/h3/image-to-video when animating a source image, and minimax/h3/reference-to-video when visual references should guide subject identity, appearance, or style.
Does MiniMax H3 support image-to-video generation?
Yes. The minimax/h3/image-to-video endpoint generates video from a still image, making it suitable for animating product photos, campaign art, character images, concept frames, and other existing visuals.
What is MiniMax H3 reference-to-video?
MiniMax H3 reference-to-video generates a new video while using visual references to guide the subject, character identity, style, or scene language. It is the preferred H3 endpoint for recurring characters and brand-consistent creative workflows.
How do I use the MiniMax H3 API on WaveSpeedAI?
Choose the H3 endpoint that matches your source, authenticate with your WaveSpeedAI API key, submit the model input as JSON, and use the returned prediction ID or result URL to retrieve the generated video. The endpoint cards and code samples on this page provide the current request schema.
What is the MiniMax H3 API?
MiniMax H3 is a MiniMax video generation model exposed as a REST API on WaveSpeedAI. MiniMax H3 API for cinematic text-to-video, image-to-video, and reference-to-video generation with strong motion quality, subject consistency, and scene coherence. Build social videos, ads, product showcases, character scenes, and scalable AI video workflows through WaveSpeedAI. You can call it programmatically or try it from the playground linked above.
How do I call the MiniMax H3 API?
Sign up for a WaveSpeedAI account, copy your API key from /accesskey, then POST to https://api.wavespeed.ai/api/v3/minimax/h3/text-to-video with your input as JSON. The endpoint returns a prediction id. Poll the result endpoint starting around every 2 seconds, increase the interval for long-running tasks, and stop on any terminal status. Production-oriented Python / Node.js / cURL examples are above.
How much does the MiniMax H3 API cost?
Pricing scales with the parameters you set in the playground (resolution, duration, output count, references). The exact cost is shown live next to the Generate button before you submit. New accounts come with free starter credits.
Which MiniMax H3 variants are available?
Browse the model catalog at /models to see all available MiniMax H3 variants.
Can I use MiniMax H3 outputs commercially?
Commercial usage rights follow the MiniMax model license. Most MiniMax models permit commercial output use; see each model's playground page for the specific license summary, and WaveSpeedAI's Terms of Service for platform-level conditions.
Why use MiniMax H3 on WaveSpeedAI instead of going direct?
One API key + one billing account across MiniMax H3 AND 1,000+ other AI models from other providers. No per-vendor SDK setup, no separate rate-limit envelopes, no rewrite-per-vendor integration code. Pricing is typically at parity with or below MiniMax's direct API.
About MiniMax
The team behind MiniMax H3 and the broader MiniMax model lineup on WaveSpeedAI.
MiniMax is a Chinese AI lab known for Hailuo video generation and speech models. Hailuo 2.3 ships physics-aware text-to-video and image-to-video with Standard, Pro, and Fast tiers — positioned around 2.5× efficiency and strong complex-instruction accuracy for creator and marketing workflows.
Start building with MiniMax H3 on WaveSpeedAI
Free starter credits on signup. One API key across 1,000+ AI models from MiniMax and every other provider.