MiniMax H3 video model debuts on Together AI with 2K output and native stereo audio
togethercompute · x · 2026-09-16
MiniMax H3, the third generation of its video line, is now live on Together AI with open weights under the Community License. The omni-modal model accepts up to 9 images, 3 videos, and 3 audio clips as unified context, with reference/editing relationships described in natural language — replacing separate text-to-video, first-last-frame, and editing expert models. It generates 4–15s clips at up to 2K via in-context regeneration, and jointly models native 32kHz stereo audio (voice, SFX, music) with stable dialogue in 11 languages. MiniMax says early testing shows it ready for commercial content creation.
Related event: MiniMax H3 Omni-Modal Video Model Lands on Together AI(2 posts)→
More from Multimodal
- Audio8-TTS-Realistic-Air, a zero-shot voice-cloning TTS model, trends on Hugging Face — Edge0 · 2026-11-02
- Mistral explains why its model-built solver is a first step toward new drugs and batteries — MistralAI · 2026-11-02
- Mistral Large 4 writes a quantum chemistry SCF solver from scratch with one short prompt — MistralAI · 2026-11-02
- Gorgeous dithered pixel art collage serves up a 'great morning' — philipvollet · 2026-11-02
- Clown Prophecy moves to new repo, now uses first clip as reference to fix H3 quality drift — reeight · 2026-11-02
- A 25-year design veteran asks: who still mixes pro software with AI tools? — Opening-Knee-5913 · 2026-11-02