China’s MiniMax Releases H3 Video Model
H3 can generate 15-second 2K clips with stereo sound, and MiniMax said it will release the model weights within days.
7 Articles
7 Articles
China’s MiniMax and ByteDance Release Dueling AI Video Models
MiniMax Group Inc. and ByteDance Ltd. released updates to their dueling AI video generation models within hours of each other, underscoring the aggressive advances that have made China the leader over the US in the burgeoning arena.
MiniMax Releases MiniMax H3: An Omni-Modal Video Model That Generates 15-Second 2K Clips With Native Stereo Audio
MiniMax releases MiniMax H3, a general-purpose multimodal generation model. MiniMax H3 is not a text-to-video model with add-ons. MiniMax describes it as a general-purpose multimodal generation model that reads text, images, video, and audio as one unified context and returns video with native stereo sound. The mains specs include: 2K output, 4–15 seconds, integer durations only. Previous video stacks split into text-to-video, image-to-video, fi…
MiniMax Releases Open-Source Full-Modal Model H3: Video Editing Ranks No.1 Globally With Pricing Cut to One-Third of Competitors
MiniMax released its first open-source multi-modal generation model H3 on July 31, a price bomb in AI video generation. It supports 2K output and 15-second audio-visual generation, ranking first globally in video editing on the Artificial Analysis leaderboard. At 0.8 yuan per second, pricing is one-third of comparable flagships. Shares rose 14% on the announcement, reversing a decline, as investors backed the cost-control strategy. H3 understand…
MiniMax H3: New Multimodal AI Sparks Bold Buzz
Chinese AI company MiniMax has started teasing what seems to be its next flagship artificial intelligence model, MiniMax H3. In a post shared on X (formerly Twitter), the company kept it short, just writing, “It’s coming,” then adding the hashtag #MiniMaxH3, plus a short teaser video that repeats the vibe, “The world is multimodal.” Even […]
Coverage Details
Bias Distribution
- 100% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
