MiniMax Launches H3: Native 2K Video with Stereo Audio Generation
Mobile-Pumpkin7944 · reddit · 2026-08-03
MiniMax has officially released MiniMax-H3, a general-purpose omni-modal generative system. It supports unified understanding of multimodal contexts composed of text, images, video, and audio.
For generation, H3 can output up to 2K resolution video with native stereo audio for up to 15 seconds. Thanks to its task-generalization-oriented design, it possesses strong multimodal instruction-following capabilities right from the pre-training stage.
Related event: MiniMax open-sources 33B multimodal video model H3(35 posts)→
More from Multimodal
- Fan animation made in Blender with Claude Code (Opus 5.5) — Icy-Appointment3583 · 2026-10-16
- Making Blender animations with Opus, Astra and Argon: how you prompt still matters — chongdashu · 2026-10-16
- Tessellation-free geometric queries on displaced surfaces cut caustics variance 10x — ssh4net · 2026-10-16
- Complex multi-link CAD mechanism finally animates, dev celebrates the breakthrough — zetalyrae · 2026-10-16
- OpenMOSS releases MOSS-Image-SFT: Apache 2.0 text-to-image model on Hugging Face — bdsqlsz · 2026-10-16
- Hyperflow open-sourced: 49 sampling steps down to 8, ~3x faster video generation — mhdfaran · 2026-10-16