[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-comfyui-day0-minimax-h3-2k-video":3,"news-related-a970be54-0603-4446-b92e-67c887c0788a":38},{"id":4,"title":5,"summary":6,"content":7,"original_url":8,"source_id":9,"tags":10,"translations":24,"news_slug":31,"published_at":32,"created_at":33,"modified_at":34,"is_published":35,"publish_type":36,"image_url":14,"view_count":37},"a970be54-0603-4446-b92e-67c887c0788a","ComfyUI Day-0 跑通 MiniMax H3:把 2K 多模态视频生成压到 RTX 3060","MiniMax 开放权重发布 H3(Hailuo 3.0),ComfyUI 在发布当天通过 66% 显存压缩把 2K 多模态视频生成跑进消费级 GPU;Artificial Analysis 榜单显示 H3 在视频编辑方向排名第一,文生视频与图生视频分列第二、第三。","# ComfyUI Day-0 跑通 MiniMax H3:把 2K 多模态视频生成压到 RTX 3060\n\n> 一段 0.5 秒决策的钩子:当一个原生支持 2K + 立体声的开源多模态视频模型在发布当天就能在 12GB 显存的 RTX 3060 上跑起来,本地化的\"剧组级 AI 视频\"第一次有了真正的桌面化路径。\n\n## 背景:从 Hailuo 到 H3,开源权重终于落地\n\n2026 年 8 月 3 日,上海稀宇科技(MiniMax)以\"开放权重\"形式正式发布第三代通用多模态生成模型 MiniMax H3(Hailuo 3.0)。官方博客将 H3 描述为\"可同时理解文本、图像、视频、音频上下文的全能型生成模型\",支持原生立体声视频生成,分辨率最高 2K、时长最高 15 秒([MiniMax Blog](https:\u002F\u002Fwww.minimax.io\u002Fblog\u002Fminimax-h3))。H3 是 MiniMax 在 Hailuo 01、Hailuo 02 之后推出的第三代视频模型,也是该公司首次以开放权重形式发布的视频模型——Hugging Face 模型卡显示权重已同步上线([Hugging Face: MiniMaxAI\u002FMiniMax-H3](https:\u002F\u002Fhuggingface.co\u002FMiniMaxAI\u002FMiniMax-H3))。\n\n第三方评测机构 Artificial Analysis 在 X 平台发布榜单:在视频编辑方向 H3 排名第一,文生视频排名第二,图生视频排名第三([Artificial Analysis on X](https:\u002F\u002Fx.com\u002FArtificialAnlys\u002Fstatus\u002F2083042088338538594)),这是\"开放权重\"模型首次在 AI 视频评测榜单上冲到榜首位置。The Decoder 8 月 3 日援引榜单数据,撰文称 MiniMax H3 为\"首个登顶 AI 视频榜单的开源模型\"([The Decoder 报道](https:\u002F\u002Fthe-decoder.com\u002Fchinas-minimax-h3-is-the-first-open-model-to-top-an-ai-video-ranking\u002F))。\n\n同日,Comfy Org 在 Substack 发布题为 \"MiniMax H3 Day-0 Support in ComfyUI\" 的工程笔记,详细披露了 H3 在 ComfyUI 0.30.0 上的本地化适配思路与显存优化路径([Comfy Org 博客](https:\u002F\u002Fblog.comfy.org\u002Fp\u002Fminimax-h3-day-0-support-in-comfyui))。文章中两个数据尤为关键:通过将约占总参数 40% 的\"调制权重\"剪枝为功能等价的查找表,叠加 int8 convrot 量化与定制推理 kernel,ComfyUI 将 H3 最小模型变体的总显存占用从全精度下的 123.6 GB 降到 42.5 GB,降幅 66%。叠加动态显存卸载后,该 2K 视频模型得以在 RTX 3060 级消费级显卡上完成本地推理。\n\n## 核心内容:ComfyUI 做了什么\n\nComfyUI 工程团队的工作可以拆成三件事,每一件都直接影响\"能不能在本地用\":\n\n- **多模态上下文理解**(Multimodal context understanding)被官方作为\"主推能力\"列出。H3 同时接收图片、音频与视频,围绕\"输入之间的关系\"进行跨模态推理,把\"分镜 + 配音 + 参考视频\"塞进一个 prompt,模型自己处理\"它们怎么对得上\"。这与文生视频\u002F图生视频分立的多模型流水线不同——后者把不同模态的生成、参考、编辑拆给独立模型拼接。\n- **原生立体声**(Native stereo audio)是模型属性而非后处理:每一段音轨从模型一次前向中直接输出,而不是由独立 TTS 或音频模型后拼上去。The Decoder 报道也指出,H3 单次 prompt 最多可吃下 9 张参考图、3 段参考视频与 3 段参考音频([Hugging Face 模型卡](https:\u002F\u002Fhuggingface.co\u002FMiniMaxAI\u002FMiniMax-H3))。\n- **编辑与动作迁移**支持\"参考视频供给动作 + 风格来自他处\"的工作流,结合 in-place 编辑可迭代某一镜头的形态。这对图生视频\u002F视频生视频场景尤其重要——之前的方案大多只能整段重生成。\n\nComfyUI 把 H3 接进工作流后,三个官方模板可以直接下载:文生视频(T2V)、图生视频(I2V)、参考生视频(R2V)([Comfy-Org\u002Fworkflow_templates](https:\u002F\u002Fgithub.com\u002FComfy-Org\u002Fworkflow_templates\u002Fblob\u002Fmain\u002Ftemplates\u002Fvideo_minimax_h3_i2v.json) 等)。权重托管在 Comfy-Org\u002FMiniMax-H3,本地运行只需更新到 ComfyUI 0.30.0,按工作流指引下载模型即可。\n\n值得注意的是,ComfyUI 这条工作流跑出来的最高分辨率是 768p,The Decoder 报道同样提到\"H3 在 ComfyUI 本地推理上限为 768p\"——2K 输出依赖未开源的 H3-Context-IR 模块,该模块负责把 prompt 与参考物料翻译成结构化中间表示。开放权重版本目前只覆盖到生成侧,留出了 prompt 工程与微调空间给社区,但没有把\"2K\"和\"完整上下文编排\"两件事一起送出来。\n\n## 个人评论:H3 的工程意义大于\"又一个视频模型\"\n\n把 H3 摆在 2026 年下半年的视频生成版图里看,真正有意思的是三件事:\n\n第一,**\"原生多模态\"从概念走向工作流**。过去一年大多数视频模型的卖点是\"画面好、动作稳\",参考理解和声音通常靠外挂。H3 把\"输入四种模态 + 一次前向输出带立体声视频\"作为架构目标,辅以 R2V、in-place 编辑等控制面。是否真的\"原生\",还要看社区基准与微调实验;但 ComfyUI 这次在 Day-0 把多模态上下文工程跑通,说明 API 上是自洽的。\n\n第二,**开放权重的 2K 视频模型第一次有\"桌面化\"路径**。ComfyUI 公布的 66% 显存下降 + 动态卸载,把\"消费级 GPU 跑 2K 多模态视频\"从口号推到可执行——前提是用户接受 768p 的本地工作流上限。ByteDance 同日发布的闭源 Seedance 2.5 走 30 秒长片段路线,形成\"开源短而多模态 vs 闭源长而单源\"的对照。\n\n第三,**中文厂商把\"多模态权重+基础设施\"打包输出**。从 7 月 16 日 Moonshot 推出 2.8T 参数 Kimi K3(被业界称为\"首个 3T 级开源模型\"([openlm.ai\u002FKimi K3](https:\u002F\u002Fopenlm.ai\u002Fkimi-k3\u002F))),到 7 月 19 日阿里以\"2.4T 参数\"旗号开放 Qwen3.8-Max 预览,再到 MiniMax H3 以 33B 参数的开放权重形态登顶 AI 视频榜单——8 月的几次发布把\"中文厂全栈开源\"从 LLM 延伸到了多模态生成。\n\n## 所以呢:对从业者意味着什么\n\n如果你在做 AI 视频产品或者本地化创作工具:\n\n- 立刻把 ComfyUI 升到 0.30.0,把三个 H3 模板跑一遍。文生\u002F图生\u002F参考生三条工作流覆盖了 80% 的\"非交互式\"用法,可直接对比 Veo、Kling、Runway 的同 prompt 输出,验证 H3 在哪些镜头类型上仍有差距。\n- 想做\"2K 商业交付\"的项目,目前还得等 MiniMax 把 2K 推理模块和 H3-Context-IR 放出来;短期内最稳的路径是 768p 出样片 + 后期放大。\n- 在 H3 的 33B 多模态架构上做领域微调(自家 IP、特定风格、特定人脸)具备可行性,但商用许可明确划了线:年收入 2,000 万美元以下的公司可用,这条线以上的厂商需要单独谈。\n\n2026 年下半年的视频生成赛道,已经不再是\"谁更长、谁更清晰\"的一维竞争。\"多模态原生 + 开源权重 + 桌面可跑\"是 H3 给出的三段式答案——下一轮竞争点会落在\"开放权重的 2K + 完整编排模块\"以及\"消费级显卡跑得动的长视频\"两个交集上。\n\n## 参考\n\n- MiniMax H3 官方博客:\u003Chttps:\u002F\u002Fwww.minimax.io\u002Fblog\u002Fminimax-h3>\n- Hugging Face 模型卡:\u003Chttps:\u002F\u002Fhuggingface.co\u002FMiniMaxAI\u002FMiniMax-H3>\n- ComfyUI Day-0 工程笔记:\u003Chttps:\u002F\u002Fblog.comfy.org\u002Fp\u002Fminimax-h3-day-0-support-in-comfyui>\n- Comfy-Org 工作流模板:\u003Chttps:\u002F\u002Fgithub.com\u002FComfy-Org\u002Fworkflow_templates\u002Fblob\u002Fmain\u002Ftemplates\u002Fvideo_minimax_h3_i2v.json>\n- The Decoder 报道:\u003Chttps:\u002F\u002Fthe-decoder.com\u002Fchinas-minimax-h3-is-the-first-open-model-to-top-an-ai-video-ranking\u002F>\n- Artificial Analysis 榜单:\u003Chttps:\u002F\u002Fx.com\u002FArtificialAnlys\u002Fstatus\u002F2083042088338538594>\n- lmmarketcap 模型动态:\u003Chttps:\u002F\u002Flmmarketcap.com\u002Fllm-updates>","https:\u002F\u002Fblog.comfy.org\u002Fp\u002Fminimax-h3-day-0-support-in-comfyui","14a63ee2-30f2-4e70-bb8a-eea116cebe84",[11,15,18,21],{"id":12,"name":13,"slug":13,"description":14,"color":14},"fca9258a-9430-455a-b95d-b9fae5e373a8","ai-inference",null,{"id":16,"name":17,"slug":17,"description":14,"color":14},"499f4b56-819d-49a3-9609-33e775143b86","multimodal",{"id":19,"name":20,"slug":20,"description":14,"color":14},"b9bd9039-fcdb-41a8-b85b-fc1587def2b9","open-source",{"id":22,"name":23,"slug":23,"description":14,"color":14},"ebe5dcd1-46b1-4298-b8c2-8e0e2f456e56","video-generation",[25],{"id":26,"lang":27,"title":28,"summary":29,"content":30},"1c3a10f9-19c0-4a38-8989-e3f39d6c289c","en","ComfyUI runs MiniMax H3 on day one: 2K video generation on RTX 3060","On 3 August 2026 MiniMax released the open-weights MiniMax H3 (Hailuo 3.0) omni-modal video model, and ComfyUI shipped Day-0 support that compresses the 2K inference memory footprint by 66% so the model runs locally on a 12 GB RTX 3060. The Artificial Analysis leaderboard puts H3 at #1 in video editing, #2 in text-to-video, and #3 in image-to-video.","# ComfyUI Day-0 support for MiniMax H3: a 2K omni-modal video model that fits on an RTX 3060\n\n> Hook: On the same day a 2K-resolution, stereo-audio, open-weights omni-modal video model ships, ComfyUI lands a working local-inference path on a 12 GB RTX 3060. That is the first real desktop route for \"production-grade\" AI video.\n\n## Background: from Hailuo to H3, the open-weights moment\n\nOn 3 August 2026, Shanghai-based MiniMax (MiniMax Group Inc.) officially released the third-generation omni-modal generation model **MiniMax H3** (a.k.a. Hailuo 3.0) as an open-weights release. MiniMax's blog describes H3 as \"a general-purpose omni-modal generation model that can jointly understand multimodal contexts spanning text, images, video, and audio\" and generate video with \"native stereo audio at up to 2K resolution and 15 seconds in length\" ([MiniMax blog](https:\u002F\u002Fwww.minimax.io\u002Fblog\u002Fminimax-h3)). H3 is the third video model in the Hailuo lineage (after Hailuo 01 and Hailuo 02) and the first that MiniMax has released with open weights. The model card is live on Hugging Face ([Hugging Face: MiniMaxAI\u002FMiniMax-H3](https:\u002F\u002Fhuggingface.co\u002FMiniMaxAI\u002FMiniMax-H3)).\n\nThe third-party evaluation house Artificial Analysis posted a leaderboard on X showing H3 ranking #1 in Video Editing, #2 in Text-to-Video, and #3 in Image-to-Video ([Artificial Analysis on X](https:\u002F\u002Fx.com\u002FArtificialAnlys\u002Fstatus\u002F2083042088338538594)). The Decoder wrote on 3 August 2026 that MiniMax H3 was \"the first open model to top an AI video ranking\" ([The Decoder](https:\u002F\u002Fthe-decoder.com\u002Fchinas-minimax-h3-is-the-first-open-model-to-top-an-ai-video-ranking\u002F)).\n\nThe same day, Comfy Org published \"MiniMax H3 Day-0 Support in ComfyUI\" on its Substack newsletter, detailing the engineering work that brought H3 into ComfyUI 0.30.0 and the optimizations that made local inference feasible on consumer hardware ([Comfy Org blog](https:\u002F\u002Fblog.comfy.org\u002Fp\u002Fminimax-h3-day-0-support-in-comfyui)). Two numbers from the post matter most: by pruning roughly 40% of the model parameters (the \"modulation weights\") into a functionally equivalent lookup table, layering an int8 convrot quantization on top, and shipping custom inference kernels, the Comfy team pushed the total memory footprint of the smallest H3 variants from 123.6 GB in full precision down to 42.5 GB — a 66% reduction. Combined with dynamic VRAM offloading, the result is a 2K-class omni-modal video model that runs locally on an RTX 3060.\n\n## What ComfyUI actually did\n\nThe ComfyUI integration can be broken into three engineering moves, each of which directly affects \"can I run this at home\":\n\n- **Multimodal context understanding** is the headline capability. H3 takes images, audio, and video in one prompt and resolves them against a description of how the inputs relate, collapsing what would otherwise be five separate tasks (script + image references + audio + reference video + edit pass) into a single model call. The implication: H3 is not a T2V model that happens to accept an image, it is a different architectural choice about where the cross-modal work happens.\n- **Native stereo audio** is a model property, not a post-process. Every audio output is generated in stereo from the same forward pass that produces the video — there is no separate TTS or audio model bolted on afterwards. The Decoder notes that a single H3 prompt can include up to 9 reference images, 3 reference video clips, and 3 reference audio clips ([Hugging Face model card](https:\u002F\u002Fhuggingface.co\u002FMiniMaxAI\u002FMiniMax-H3)).\n- **Editing and motion transfer** support a reference-video-supplies-motion, style-comes-from-elsewhere workflow, and combined with in-place editing that means a single shot can be iterated without regenerating the whole clip. For a graph-style pipeline (ComfyUI's natural habitat), this is the capability that changes how workflows are built.\n\nComfyUI ships three official workflow templates: text-to-video (T2V), image-to-video (I2V), and reference-to-video (R2V) ([Comfy-Org workflow_templates](https:\u002F\u002Fgithub.com\u002FComfy-Org\u002Fworkflow_templates\u002Fblob\u002Fmain\u002Ftemplates\u002Fvideo_minimax_h3_i2v.json)). Weights are hosted at `Comfy-Org\u002FMiniMax-H3`; running locally requires ComfyUI 0.30.0 or later plus the model files in the standard ComfyUI directories.\n\nA practical limit to be aware of: ComfyUI's local H3 workflow caps at 768p. The Decoder reports the same ceiling. The 2K path depends on an **H3-Context-IR** module that turns prompts and reference material into a structured intermediate representation, and that module is not included in the open-weights release. The community is expected to assemble the context side themselves using MiniMax's published prompting guides.\n\n## Opinion: H3's engineering impact outranks \"yet another video model\"\n\nH3 lands at an interesting moment in the 2026 video-generation landscape, and three things stand out:\n\n**1. \"Native omni-modal\" is moving from concept to working pipeline.** Most video models in the past 12 months sold themselves on visual quality and motion stability, with audio and reference understanding handled by separate models. H3 takes four input modalities in one forward pass and emits synchronized stereo video. Whether the underlying mechanism is \"truly native\" or a tightly-coupled multi-encoder stack is something the open-weights release will let the community verify. What is clear from the ComfyUI Day-0 integration is that the API surface is self-consistent enough to build a real workflow on.\n\n**2. Open-weights 2K video has its first real \"desktop\" path.** ComfyUI's 66% memory reduction plus dynamic VRAM offloading is what moves \"consumer GPU running a 2K multi-modal video model\" from marketing slide to executable. The local ceiling is 768p, so production 2K output still requires either the closed 2K module from MiniMax or an upscale pass — but the bottleneck is no longer \"you need eight A100s to try the model.\" The same day, ByteDance released the closed **Seedance 2.5** with 30-second clips and built-in audio, which is a different point on the design space (\"long + closed\" vs. \"open + multi-modal + short\").\n\n**3. Chinese vendors are bundling \"multi-modal weights + infrastructure\" together.** The pattern repeats: 16 July 2026 saw Moonshot's Kimi K3 launch as a 2.8T-parameter open-weights model (\"the world's first open 3T-class model\" per Moonshot's own framing on [openlm.ai](https:\u002F\u002Fopenlm.ai\u002Fkimi-k3\u002F)); 19 July 2026 saw Alibaba open-weight the Qwen3.8-Max preview at 2.4T parameters ([Alibaba Cloud blog](https:\u002F\u002Fwww.alibabacloud.com\u002Fblog\u002Falibaba-cloud-unveils-agent-native-innovations-at-waic-2026_603377)); and on 3 August 2026 MiniMax H3 took the AI video leaderboard with a 33B open-weights release. The \"Chinese stack goes fully open\" story that started in LLMs is now visibly extending to multi-modal generation.\n\n## So what: what practitioners should actually do\n\nFor anyone shipping AI video products or building local creative tooling:\n\n- **Update ComfyUI to 0.30.0 and run all three H3 templates today.** T2V, I2V, and R2V cover roughly 80% of non-interactive use cases, and side-by-side comparison against Veo \u002F Kling \u002F Runway on the same prompt is the fastest way to locate where H3 still has visible gaps.\n- **For 2K commercial deliverables, wait for the missing pieces.** The 2K inference module and the H3-Context-IR are still closed. Until MiniMax releases them, the practical path is 768p for offline previews plus a separate upscale pass.\n- **Domain fine-tuning is feasible but commercially gated.** Fine-tuning H3 on a specific IP, character, or visual style is supported by the open weights, but the commercial-use license restricts it to companies under US$20M in annual revenue. Larger companies need to negotiate separately.\n\nThe second half of 2026 is no longer a one-dimensional race about \"who is longer, who is sharper.\" H3's three-part answer is \"omni-modal native + open weights + desktop-runnable.\" The next frontier is the intersection of \"open-weights 2K + complete context module\" and \"consumer-GPU long video.\"\n\n## References\n\n- MiniMax H3 blog: \u003Chttps:\u002F\u002Fwww.minimax.io\u002Fblog\u002Fminimax-h3>\n- Hugging Face model card: \u003Chttps:\u002F\u002Fhuggingface.co\u002FMiniMaxAI\u002FMiniMax-H3>\n- ComfyUI Day-0 engineering note: \u003Chttps:\u002F\u002Fblog.comfy.org\u002Fp\u002Fminimax-h3-day-0-support-in-comfyui>\n- Comfy-Org workflow templates: \u003Chttps:\u002F\u002Fgithub.com\u002FComfy-Org\u002Fworkflow_templates\u002Fblob\u002Fmain\u002Ftemplates\u002Fvideo_minimax_h3_i2v.json>\n- The Decoder coverage: \u003Chttps:\u002F\u002Fthe-decoder.com\u002Fchinas-minimax-h3-is-the-first-open-model-to-top-an-ai-video-ranking\u002F>\n- Artificial Analysis leaderboard: \u003Chttps:\u002F\u002Fx.com\u002FArtificialAnlys\u002Fstatus\u002F2083042088338538594>\n- lmmarketcap model timeline: \u003Chttps:\u002F\u002Flmmarketcap.com\u002Fllm-updates>\n- Kimi K3 framing: \u003Chttps:\u002F\u002Fopenlm.ai\u002Fkimi-k3\u002F>\n- Qwen3.8-Max preview: \u003Chttps:\u002F\u002Fwww.alibabacloud.com\u002Fblog\u002Falibaba-cloud-unveils-agent-native-innovations-at-waic-2026_603377>","comfyui-day0-minimax-h3-2k-video","2026-08-03T04:00:00Z","2026-08-03T14:43:58.146726Z","2026-08-03T14:43:58.146736Z",true,"agent",1374,{"items":39},[40,45,50,55,60,65],{"id":41,"title":42,"news_slug":43,"published_at":44},"aad00b18-d354-48b5-ad21-62b53150b8c6","MiniMax H3 开源实测:你下载的权重,和 API 里跑的不是同一个模型","minimax-h3-local-vs-api-gap","2026-08-15T17:07:24+00:00",{"id":46,"title":47,"news_slug":48,"published_at":49},"bcdc10bc-2f08-4c39-8ffa-e7e34041c112","京东开源 JoyAI-Video-Edit:用 16B 多模态扩散 Transformer 把视频编辑推进「边播边改」实时流时代","jd-joyai-video-edit-real-time-streaming","2026-08-05T03:00:00+00:00",{"id":51,"title":52,"news_slug":53,"published_at":54},"b4214f43-353e-42e3-b48e-92dd4fc64290","京东开源 EchoWM 全模态世界模型:720p 音画同步,能跟着你走","jd-echowm-omnimodal-world-model","2026-08-25T23:10:00+00:00",{"id":56,"title":57,"news_slug":58,"published_at":59},"2874a2e5-beae-4627-8f6f-a34cf2cc8d7a","一段随手拍视频直出4D人体:4DAnyone用RCP+TCR破解多视角一致性,代码权重全开源","4danyone-monocular-video-4d-human","2026-08-20T17:59:53+00:00",{"id":61,"title":62,"news_slug":63,"published_at":64},"b95b93e8-294a-4c5b-b53d-ce6ea07c1519","SemComp-Bench 登顶 Hugging Face 日榜:视频生成开始考「任务做没做成」","semcomp-bench-video-task-completion","2026-08-20T13:30:00+00:00",{"id":66,"title":67,"news_slug":68,"published_at":69},"6f9e9f94-9dcc-4c6c-b254-6c5d0fe8ed37","京东开源 JoyAI-Video-Edit:16B 多模态扩散 Transformer 把视频编辑推进「边播边改」实时流时代","jd-joyai-video-edit-realtime-diffusion","2026-08-10T00:00:00+00:00"]