核心信息
MiniMax H3 是一款开源权重的视频生成模型，支持原生立体声和多模态参考控制。社区项目 FastH3 与 Sol-H3 正显著提升其运行速度和部署便利性。
要点
- FastH3（FastVideo、Nuva Lab 与 NVIDIA 联合开发）实现 4 步蒸馏，可在 DGX Spark 和 Apple Silicon 上运行。
- Sol-H3（NVIDIA SANA 团队）采用 H3 + LTX-2.5 两阶段管线，并已在 DGX Spark 上适配，可扩展至 8×B300。
MiniMax H3：开源权重视频生成模型获得社区加速
原始推文
Open weights. Shared progress. MiniMax H3 is moving fast. We built MiniMax H3 for video generation with native stereo audio and multimodal reference control. The open-source community is making that capability faster, more accessible, and easier to build on. Recent highlights: • FastH3 — FastVideo, Nuva Lab and NVIDIA: 4-step distillation, now running on DGX Spark and Apple Silicon. • Sol-H3 — NVIDIA’s SANA team: now on DGX Spark with a two-stage H3 + LTX-2.5 pipeline. On 8×B300, the team reports 15 seconds of 768p video + audio in 6.6 seconds of warm inference.* • VDN — Haocheng Xi and the OpenVDN team: rethinking attention for faster H3 inference, with weights, training and inference code released. • PDD — NVIDIA’s distillation method, brought to H3 by Alibaba PAI as 8-step Acc-LoRAs, now supported in ComfyUI. • LightX2V — 4- and 8-step Turbo LoRAs, with workflows for text, image and reference-conditioned video + audio. Behind every release are people training, optimizing, quantizing, testing and sharing. To those teams—and everyone building the nodes, ports and workflows that make MiniMax H3 usable—thank you. Powerful models go further when we build together. Keep pushing MiniMax H3. Explore the ecosystem: https://t.co/F1S5IjUNi1 *Sol-H3 timing excludes model loading, compilation and MP4 encoding.