FastVideo 发布 FastH3-Omni 模型
FastVideo team excited to announce FastH3-Omni! 🚀🚀
FastVideo 团队激动宣布 FastH3-Omni！🚀🚀 与 Nuva Lab @nuvalab 和 Reactor @reactorworld 合作进行后训练！ 最终的 FastH3-Omni 模型将于 10 月 20 日完全开源，同时提供训练代码，并在 DGX spark 和 RTX GPU 上提供完整支持！ 请查看 @reactorworld 的 API 访问，今天即可使用！
查看原帖Introducing FastH3-Omni. The fastest and highest-quality open-source version of realtime H3 by @MiniMax_AI, post-trained by FastVideo @haoailab with @nuvalab. Exclusive preview on the Reactor API until 10.20.2026. • First and last frame conditioning • Audio input + strong dialogue • Up to 3 reference images • Frontier-level performance
正文
当前显示原文，中文全文译文尚未就绪。
FastVideo team excited to announce FastH3-Omni! 🚀🚀 Postrained in collaboration with Nuva Lab @nuvalab and Reactor @reactorworld!
The final FastH3-Omni model will be fully open sourced on October 20, along with training code and full support on DGX spark, and RTX GPUs!
Please checkout @reactorworld's API access, available today!