Newsletter
Join the AIhubs Community
Get weekly updates on the latest AI tools, resources, and insights delivered straight to your inbox
Run MiniMax H3 text-to-video, image-to-video, and reference-to-video models through a production-ready API with native audio and flexible video durations.
WaveSpeedAI provides API access to the MiniMax H3 model family for text-to-video, image-to-video, and multimodal reference-to-video generation. Available capabilities include native stereo audio, 5–15 second video generation, flexible aspect ratios, subject consistency, and scene continuity.
MiniMax H3 on WaveSpeedAI is a collection of AI video generation models for text-to-video, image-to-video, and reference-to-video workflows. Developers and creators can generate videos from prompts, animate still images, or use image, video, and audio references to guide subject identity, movement, style, and scene continuity. The collection includes ready-to-use REST APIs, per-generation pricing, native audio support, and output options ranging from 480p and 768p to 2K, depending on the selected model.
