MiniMax H3 AMA 回顾：开源计划与未来路线图
Thank you to everyone who joined our Reddit AMA! The community energy was incred…
Thank you to everyone who joined our Reddit AMA! The community energy was incredible, and we loved diving deep into the architecture, workflows, and the future of MiniMax H3.🩵
感谢所有加入我们 Reddit AMA 的人！社区的能量令人难以置信，我们很喜欢深入探讨 MiniMax H3 的架构、工作流程和未来。🩵
For those who missed it, here is a full recap of what is shipping next and YES WE WILL KEEP OPEN UNTIL AGI ARRIVES.
对于那些错过的人，这里是接下来发布内容的完整回顾，是的，我们将保持开放直到 AGI 到来。
🔥 The Open-Source Mission As copyright matters settle, transitioning to an Apache-2.0 license is on the table. We also owe you the technical details - a comprehensive technical report on H3's development is in the works and will be published soon~
🔥 开源使命 随着版权问题的解决，过渡到 Apache-2.0 许可证已提上日程。我们也欠你们技术细节——一份关于 H3 开发的全面技术报告正在编写中，并将很快发布~
🎥 Video & H3-Regenerate-2K We are planning to open-source H3-Regenerate-2K. To be clear, this is a dedicated latent-space DiT regeneration model - not just a base checkpoint rerun and not a pixel upscaler. We are currently tuning its efficiency and quality so you can run it locally.
🎥 视频与 H3-Regenerate-2K 我们计划开源 H3-Regenerate-2K。需要明确的是，这是一个专用的潜在空间 DiT 再生模型——不是简单的基础检查点重跑，也不是像素级超分辨率器。我们目前正在调整其效率和质量，以便你可以在本地运行它。
⚡ Architecture, Speed & Sparse Attention Sparse Attention: Our sparse attention is MoBA-style, train-aware block selection. Expect a relatively conservative reference implementation in the near term with the goal of zero perceptible quality loss. Let’s build device-specific speedups together!
⚡ 架构、速度与稀疏注意力 稀疏注意力：我们的稀疏注意力是 MoBA 风格的，基于训练感知的块选择。预计在短期内会有一个相对保守的参考实现，目标是零可感知的质量损失。让我们一起构建设备特定的加速！
Low-Step Variants: The released checkpoint is already CFG-distilled. While we don't have a near-term commitment just yet, a 4-NFE / 8-NFE variant is under active consideration. In the meantime, huge shoutout to the community - the Turbo LoRAs you've built are genuinely fantastic.
低步数变体：发布的检查点已经过 CFG 蒸馏。虽然我们目前还没有近期承诺，但 4-NFE / 8-NFE 变体正在积极考虑中。与此同时，非常感谢社区——你们构建的 Turbo LoRA 确实非常出色。
🖼️ Unified Image Generation & Editing Single-frame image generation was our one of most-asked topics (193 upvotes!). The answer is YES: we plan to open-source a unified text-to-image and general image-editing model derived directly from the H3 lineage. It is currently in post-training refinement.
🖼️ 统一图像生成与编辑 单帧图像生成是我们被问得最多的话题之一（193 个赞！）。答案是肯定的：我们计划开源一个直接源自 H3 系列的统一文本到图像和通用图像编辑模型。目前正在进行后训练优化。
⏱️ Pushing Limits: 60-Second Workflows For longer videos, Ref2VA supports continuation (just feed the previous clip as the reference). And to the Redditor who chained together that 60-second workflow - great find! That capability is real and was retained from our pretraining phase~🫡
⏱️ 突破极限：60 秒工作流程 对于更长的视频，Ref2VA 支持续接（只需将前一个片段作为参考输入）。对于那位 Reddit 用户，你串联起了那个 60 秒的工作流程——发现得真好！该功能是真实的，并且是从我们的预训练阶段保留下来的~🫡
We can't wait to see what you build next. Let's keep pushing the boundaries of open AI together. 🚀
我们迫不及待地想看到你们接下来构建的内容。让我们一起推动开放 AI 的边界。🚀
For details👇
详情👇
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力