Overview
MiniMax-H3-Max is a high-speed video generation model post-trained by fal.ai from a MiniMax base video model and optimized for high-speed generation. It supports text-to-video and image-to-video, including first and last frame input, with a maximum resolution of 768P and clips up to 15 seconds.
It accepts text and image input and returns video. Billing is per second, not per token.
Capabilities
Two generation modes are documented. Text-to-video builds a clip from a written description, and image-to-video starts from a picture, including a first frame and a last frame so that the clip moves between two images.
Supported tasks: text-to-video, image-to-video and first and last frame. Resolutions: 480P and 768P, with a maximum resolution of 768P. Clips run up to 15 seconds. Frame rate: 24 fps.
Pricing
MiniMax-H3-Max costs $0.05 per second of generated video. It is billed per second of video output rather than per token.
MiniMax-H3-Max is available through RealRelay's OpenAI-compatible API.
Use cases
Use MiniMax-H3-Max for quick text-to-video drafts, such as ad concepts, social clips and storyboard previews, where fast generation lets you iterate on a prompt many times.
Use image-to-video with a first and last frame to animate between two key visuals, for example a product shot that transitions into a lifestyle scene.
What is MiniMax-H3-Max?
MiniMax-H3-Max is a high-speed video generation model post-trained by fal.ai from a MiniMax base video model and optimized for high-speed generation. It supports text-to-video and image-to-video, including first and last frame input, with a maximum resolution of 768P and clips up to 15 seconds.
How much does MiniMax-H3-Max cost?
MiniMax-H3-Max costs $0.05 per second of generated video.
What resolutions and durations does MiniMax-H3-Max support?
MiniMax-H3-Max accepts text and image input and returns video. Supported tasks: text-to-video, image-to-video and first and last frame. Resolutions: 480P and 768P, with a maximum resolution of 768P. Clips run up to 15 seconds. Frame rate: 24 fps.
What is MiniMax-H3-Max used for?
Use MiniMax-H3-Max for quick text-to-video drafts, such as ad concepts, social clips and storyboard previews, where fast generation lets you iterate on a prompt many times.
How do I call MiniMax-H3-Max through the API?
Call MiniMax-H3-Max through RealRelay's OpenAI-compatible API by setting the model field to minimax/MiniMax-H3-Max.