MiniMax H3 Video Model Open-Sourced with 768p Local Generation
KuCoinFlashShare
On-chain news from MetaEra reveals that on August 3, MiniMax officially released the weights for its H3 video model. The launch includes two H3-Base variants—one optimized for text-to-video and first-last-frame generation, and another for image, video, and audio input. Each task is handled by a single model. H3-Base, a 33-billion-parameter dense system, supports 4–15 second, stereo 768p local outputs using SGLang, vLLM, Diffusers, and ComfyUI. The full H3 suite, including H3-Context-IR and H3-Regenerate-2K, remains proprietary and requires access via the MiniMax API for 2K results. Licensing excludes the U.S., EU, UK, and South Korea, and commercial use exceeding $200 million in revenue requires explicit approval. Token launch news continues to underscore evolving shifts in model accessibility.
Source:Show original
Disclaimer: The information on this page may have been obtained from third parties and does not necessarily reflect the views or opinions of KuCoin. This content is provided for general informational purposes only, without any representation or warranty of any kind, nor shall it be construed as financial or investment advice. KuCoin shall not be liable for any errors or omissions, or for any outcomes resulting from the use of this information.
Investments in digital assets can be risky. Please carefully evaluate the risks of a product and your risk tolerance based on your own financial circumstances. For more information, please refer to our Terms of Use and Risk Disclosure.