MiniMax Launches H3 2K Multimodal Video Model, Open Weights Promised But Not Yet Released · cho.sh
MiniMax Launches H3 2K Multimodal Video Model, Open Weights Promised But Not Yet Released
AI video model releases
MiniMax Launches H3 2K Multimodal Video Model, Open Weights Promised But Not Yet Released
MiniMax launched H3 on July 31, a multimodal video model generating clips with stereo audio at up to 2K resolution and durations of four to 15 seconds at 24 frames per second.
MiniMax says it will open H3's weights "in the coming days" but as of August 2, no public weights, license, or hardware requirements had been published.
H3 accepts text, images, video, and audio as context and supports text-to-video, image-to-video, first-and-last-frame generation, and multimodal reference generation.
MiniMax claims H3 costs less than one-third per second at 2K compared to unnamed "mainstream models," without disclosing the comparison set or pricing table.
The launch follows MiniMax closing roughly $2 billion in funding in July 2026; the company's 2025 revenue was $79 million, up from $30.5 million in 2024, with 73% from outside mainland China.
cho.sh