Tagged articles

audio-visual model

1 articles · Page 1 of 1
SuanNi
SuanNi
Aug 4, 2026 · Artificial Intelligence

MiniMax H3: Open‑Source Next‑Gen General Video Model Matching Seedance 2.0

MiniMax has open‑sourced its H3 multimodal video model, which rivals Seedance 2.0 in quality, supports text, image, video and audio inputs, generates up to 2 K stereo video on a single RTX 3060, and is built from three dedicated modules that can be run via Hugging Face checkpoints and API.

MiniMax H3audio-visual modelmultimodal AI
0 likes · 8 min read
MiniMax H3: Open‑Source Next‑Gen General Video Model Matching Seedance 2.0