MiniMax (official)@MiniMax_AI
68AI 编辑部评分,满分 100
2026-08-10 13:26· 4小时前
AI 导读

MiniMax H3 开源多模态视频生成模型在 ComfyUI 直播中亮相,支持文本、图像、视频和音频输入,可生成最长 15 秒 768p 视频片段,并原生同步生成含对话、音效和音乐的立体声音频。托管版 H3 支持最高 2K 分辨率生成,模型已优化至可在消费级 GPU 上本地部署,并提供 Context-IR 与 Regenerate-2K API 工作流。

Huge thanks to @ComfyUI for an incredible livestream!

From reference-to-video and native audio, to multi-shot commercial creation, MiniMax H3 showcased the expanding possibilities of open-weight video models. The stream also took a deep dive into quantization, offloading, Context-IR, and practical deployment on consumer-grade GPUs.

H3 is just getting started. We look forward to pushing creative boundaries alongside developers worldwide! Watch the full livestream ↓

ComfyUILivestream Update: @MiniMax_AI H3: A New Open-Weight Video Model, Live in ComfyUI MiniMax H3 is an open-weight, general-purpose multimodal video generation mode...

来源:MiniMax (official) · x.com

MiniMax (official) · @MiniMax_AI · X·2026-08-10 13:26·4小时前
AI 导读

MiniMax H3 开源多模态视频生成模型在 ComfyUI 直播中亮相,支持文本、图像、视频和音频输入,可生成最长 15 秒 768p 视频片段,并原生同步生成含对话、音效和音乐的立体声音频。托管版 H3 支持最高 2K 分辨率生成,模型已优化至可在消费级 GPU 上本地部署,并提供 Context-IR 与 Regenerate-2K API 工作流。

Huge thanks to @ComfyUI for an incredible livestream!

From reference-to-video and native audio, to multi-shot commercial creation, MiniMax H3 showcased the expanding possibilities of open-weight video models. The stream also took a deep dive into quantization, offloading, Context-IR, and practical deployment on consumer-grade GPUs.

H3 is just getting started. We look forward to pushing creative boundaries alongside developers worldwide! Watch the full livestream ↓

ComfyUILivestream Update: @MiniMax_AI H3: A New Open-Weight Video Model, Live in ComfyUI MiniMax H3 is an open-weight, general-purpose multimodal video generation mode...

来源:MiniMax (official)· x.com