MiniMax H3 Image to Video LoRA: Lora Support
Animate images with MiniMax H3 Image to Video LoRA. Custom LoRAs, native stereo audio, and 480P/768P output. Try free — pay per generation.
Animate images with MiniMax H3 Image to Video LoRA. Custom LoRAs, native stereo audio, and 480P/768P output. Try free — pay per generation.
About this model
MiniMax H3 Open Weights Image to Video LoRA animates a first-frame image, optionally with last-frame guidance, into coherent 480P or 768P videos with native stereo audio. It supports custom LoRA fine-tunes for consistent styles, character motion, and specialized effects across 3-15 second durations.
Cost analysis
| Provider | Cost | Notes |
|---|---|---|
| muapiapp | $0.03 - $0.06 / sec | Pay per second of video generated. No subscription. |
| Fal.ai | Not available | Custom LoRA support not available. |
| Replicate | Not available | Model variant not hosted. |
Pay per second of video generated. No subscription.
Custom LoRA support not available.
Model variant not hosted.
** Competitor pricing is estimated based on similar model architectures and usage tiers.
Configuration schema
| Parameter | Type | Description | Default |
|---|---|---|---|
| Prompt | string | Text description of the desired motion, scene, and soundtrack. | A cinematic ocean wave at sunrise, highly detailed |
| Start Frame Image URL | string | First-frame image URL. The output canvas follows this image's aspect ratio. | |
| Last Frame Image URL | string | Optional last-frame image URL for frame interpolation. | |
| LoRAs | array | List of LoRAs to apply (maximum 3). | - |
| Resolution | Enum (2 options) | Output video resolution. 768p is native canvas, 480p is faster. | 480p |
| Duration (seconds) | int | Output video duration in seconds. | 5 |
| Seed | int | Random seed (-1 for random) | -1 |
Text description of the desired motion, scene, and soundtrack.
A cinematic ocean wave at sunrise, highly detailedFirst-frame image URL. The output canvas follows this image's aspect ratio.
Optional last-frame image URL for frame interpolation.
List of LoRAs to apply (maximum 3).
-Output video resolution. 768p is native canvas, 480p is faster.
480pOutput video duration in seconds.
5Random seed (-1 for random)
-1Developer documentation
Provide Input Image: Upload your starting image URL (images_list). Optionally add last_image for target end frame.
Configure Prompt & LoRAs: Write a descriptive text prompt detailing camera motion and soundtrack. Pass custom LoRA adapters if desired.
Set Duration & Resolution: Pick 480p or 768p resolution and a duration between 3 and 15 seconds.
Submit Request: Send a POST request to /api/v1/minimax-h3-image-to-video-lora:
curl -X POST https://api.muapi.ai/api/v1/minimax-h3-image-to-video-lora \
-H 'Authorization: Bearer YOUR_API_KEY' \
-H 'Content-Type: application/json' \
-d '{
"prompt": "A cinematic ocean wave at sunrise, highly detailed",
"images_list": ["https://example.com/start.jpg"],
"resolution": "480p",
"duration": 5
}'
Frequently asked
It is an image-to-video AI model built on MiniMax H3 open weights supporting custom LoRA adapters and native audio generation.
It supports 480p and 768p resolutions with flexible video durations from 3 to 15 seconds.
Yes, native stereo audio is synthesized automatically alongside the generated video based on your text prompt.
Yes, passing last_image enables first-frame to last-frame video interpolation.
Billing is calculated based on video duration (in seconds) and chosen output resolution.