H3 Max
AI model · Video generation · Text → Video by fal
fal's post-trained frontier video model for high-fidelity 2K generation.
H3 Max is fal's post-trained version of MiniMax H3, specifically tuned for superior prompt adherence and aesthetic quality. It is designed for developers and creative teams who need high-throughput video generation without sacrificing output fidelity. The model supports text-to-video and image-to-video workflows, producing 2K resolution clips ranging from 5 to 15 seconds. It is delivered via fal's optimized inference stack, which provides up to 10x faster performance for diffusion models.
- Languages
- English
- Output Resolution
- Up to 2K
- Video Duration
- 5 to 15 seconds
- Input Modality
- Text, Image
- Output Modality
- Video
Overview
H3 Max is fal's post-trained version of MiniMax H3, specifically tuned for superior prompt adherence and aesthetic quality. It is designed for developers and creative teams who need high-throughput video generation without sacrificing output fidelity. The model supports text-to-video and image-to-video workflows, producing 2K resolution clips ranging from 5 to 15 seconds. It is delivered via fal's optimized inference stack, which provides up to 10x faster performance for diffusion models.
Video Generation
- High Resolution: Generates detailed video with 2K output and flexible aspect ratios.
- Flexible Duration: Supports creating clips from 5 to 15 seconds directly from text descriptions.
- Optimized Throughput: Co-optimized with a custom inference stack for higher throughput without compromising quality.
Advanced Controls
- Prompt Adherence: Tuned for stronger prompt adherence and better aesthetics compared to the base model.
Pricing
- 480p Video (Standard)
- $0.05 per second
- 768p Video (Standard)
- $0.08 per second
- 1080p Video (Standard)
- $0.16 per second
Specifications
- Output Resolution
- Up to 2K
- Video Duration
- 5 to 15 seconds
- Input Modality
- Text, Image
- Output Modality
- Video
Added by AntSpace from public sources (fal.ai).