July 31 — MiniMax announced the release of MiniMax H3, a general-purpose multimodal generative model that ingests unified text, image, video and audio context and can produce native stereo audiovisual output up to 15 seconds at 2K resolution. MiniMax said it plans to release the model weights in the coming days, subject to applicable laws and regulations.

2026-07-31

July 31 — MiniMax announced the release of MiniMax H3, a general-purpose multimodal generative model that ingests unified text, image, video and audio context and can produce native stereo audiovisual output up to 15 seconds at 2K resolution. MiniMax said it plans to release the model weights in the coming days, subject to applicable laws and regulations.