[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-zhixiang-multimodal-token-margin-world-model":3,"topics-all":36,"news-related-25eeddf5-b3d7-45b4-9725-387f60955252":55},{"id":4,"title":5,"summary":6,"content":6,"original_url":7,"source_id":8,"tags":9,"translations":23,"news_slug":29,"published_at":30,"created_at":31,"modified_at":32,"is_published":33,"publish_type":34,"image_url":13,"view_count":35},"25eeddf5-b3d7-45b4-9725-387f60955252","智象未来提出多模态模型Token利润率远超语言模型：世界模型的技术路径与商业化前景","5月19日，智象未来在首届开放日上抛出一个颇具争议的观点：多模态模型Token的毛利率，远高于大语言模型Token。这一判断背后，是这家公司从模型即产品到做Agent平台的战略转变，也折射出当下视频生成赛道的竞争正在从技术能力比拼，转向商业落地效率的较量。\n\n智象未来CEO梅涛认为，真正的世界模型需要满足四个条件：掌握物理规律、解决长时序因果推理、实现全模态交互，以及绝对的安全性。这意味着当前市面上多数自称世界模型的产品，在严格定义下都还为时过早。\n\n与李飞飞主张的生成3D物理世界和LeCun倡导的自监督预测世界不同，智象未来选择了另一条路径——从算法和架构层创新入手，率先聚焦视频、图像、3D交互等数据的生成，用低成本合成数据对抗行业数据稀缺的瓶颈。他们的核心技术架构是原生全模态Unified Transformer（UiT），做到Any to Any（任意形式输入对应任意形式输出），在统一架构中理解、生成并预测现实世界的不同状态。梅涛透露，其视频模型目前已能生成分钟级视频，技术上限可达3至5分钟乃至更长。\n\n世界模型训练需要全模态数据，成本极高且稀缺。智象未来的策略是：用视频生成模型合成大量真人实操数据，再用于训练VLA（视觉-语言-动作模型）和WAM（世界动作模型）。他们拥有20万小时有版权的视频数据。这一策略的底层逻辑是：当算法架构发生范式转变，数据的重要性会相对弱化，创业公司才有机会以少量资源比肩大厂的基模能力。\n\n从投资视角看，视频生成赛道的商业化临界点正在逼近。东方富海合伙人王兵指出，视频生成企业的商业化有三个衡量维度：团队技术背景及积累深度、人才密度与稳定性、以及资金和研发效率。梅涛则断言，多模态模型Token的毛利率远高于语言模型。随着算力成本指数级下降，叠加AI短剧、短视频、电商视频的质量今年已达到商业化水准，视频生成To B服务的毛利率正在从负转正。\n\n智象未来这番Token毛利率论断，本质上是对多模态内容商业价值的一次重新定价。语言模型输出的是文本，附加值相对单一；而视频、图像等多模态内容的应用场景更广、付费意愿更强。但挑战同样真实：版权合规问题是悬在所有视频生成公司头顶的达摩克利斯之剑。智象未来All-in正版版权数据的策略，短期内看似增加了成本，长期却是通往商业化持续性的必经之路。","https:\u002F\u002F36kr.com\u002Fp\u002F3820602297208966","5e4fd3d1-9cb4-44a6-bae5-9ffb449c05c1",[10,14,17,20],{"id":11,"name":12,"slug":12,"description":13,"color":13},"e676a5cf-1f24-472f-a765-86fa21a1bc3c","ai-model",null,{"id":15,"name":16,"slug":16,"description":13,"color":13},"0ef8513a-0a26-42f0-b6f9-5b6dadded45c","efficiency",{"id":18,"name":19,"slug":19,"description":13,"color":13},"499f4b56-819d-49a3-9609-33e775143b86","multimodal",{"id":21,"name":22,"slug":22,"description":13,"color":13},"ebe5dcd1-46b1-4298-b8c2-8e0e2f456e56","video-generation",[24],{"id":25,"lang":26,"title":27,"summary":28,"content":13},"cc8bfd14-841c-4b15-84d7-050c45d45dfb","en","Zhixiang: multimodal token margins far exceed language models","36Kr's feature on Zhixiang Future (智象未来) argues that multimodal world models have a token profit margin far exceeding that of language models. The argument: each multimodal token (e.g., a video frame) carries more information and higher value than a text token, justifying higher pricing.","zhixiang-multimodal-token-margin-world-model","2026-05-28T10:00:00Z","2026-05-28T10:05:25.422469Z","2026-08-19T02:08:40.142862Z",true,"agent",280,[37,46],{"slug":38,"tag_slug":38,"title_zh":39,"title_en":40,"intro_zh":41,"intro_en":42,"id":43,"is_active":33,"created_at":44,"modified_at":45},"ai-for-science","AI for Science 2026：从 UniPert 到 GPT-Rosalind 的硬核进化","AI for Science 2026: from UniPert to GPT-Rosalind","生命科学、化学材料、物理世界模型——AI 正在从\"语言工具\"变成\"实验伙伴\"。本专题收录 AI 在三大科学方向的关键节点：UniPert 统一基因与化学扰动空间、GPT-Rosalind 端到端生命科学推理、达摩院 AI 智能体 28 小时找到 4 种超导新材料、Anthropic Claude Science 把工作台做成标准品。","From language tool to lab partner — AI is reshaping life sciences, chemistry\u002Fmaterials, and physical world models. This topic covers the key milestones: UniPert unifying genetic-chemical perturbation spaces, GPT-Rosalind's end-to-end life-sciences reasoning, DAMO's AI agent discovering 4 superconducting materials in 28 hours, and Anthropic's Claude Science workbench going mainstream.","988a4300-5fab-41c4-b5d8-63711a2dc757","2026-09-10T01:34:15.296649Z","2026-09-10T01:34:15.296663Z",{"slug":47,"tag_slug":47,"title_zh":48,"title_en":49,"intro_zh":50,"intro_en":51,"id":52,"is_active":33,"created_at":53,"modified_at":54},"h3-series","MiniMax H3 系列：从开源权重到 35 倍吞吐","MiniMax H3 Series: from open weights to 35x throughput","MiniMax H3 自 2026 年 8 月开源以来节奏密集：官方把生成、参考与编辑收回一个模型；ComfyUI 当天压进 RTX 3060；摩尔线程 3 小时完成国产 GPU 适配；fal 后训练版把吞吐拉到 35 倍；FastH3 蒸馏再砍推理成本。本专题持续追踪 H3 的发布—开源—蒸馏—部署全链路。","Since MiniMax open-sourced H3 in August 2026 the pace has been relentless: one unified omni-modal model, same-day ComfyUI support down to an RTX 3060, a 3-hour Day-0 port to Moore Threads GPUs, fal's post-trained H3 Max at 35x throughput, and FastH3 distillation cutting inference cost further. This topic tracks the full H3 chain — release, open weights, distillation, deployment.","83ef0daa-3c31-4cb3-86ed-e5ee58654d5f","2026-09-08T07:33:19.942193Z","2026-09-08T07:33:19.942209Z",{"items":56},[57,62,67,72,77,82],{"id":58,"title":59,"news_slug":60,"published_at":61},"52e9955f-50fe-4ea1-98cc-a7a442d57b71","MilliVid：MIT 把「分层潜变量」装进视频扩散，长程一致性首次跑通长镜头","millivid-mit-hierarchical-latent-long","2026-06-11T02:00:00+00:00",{"id":63,"title":64,"news_slug":65,"published_at":66},"7d24855e-d09e-485d-a87d-77586313da28","Google Veo 3.1「Ingredients to Video」：把视频控制粒度做到最细","veo-3-1-ingredients-to-video-fine-control","2026-05-28T07:01:00+00:00",{"id":68,"title":69,"news_slug":70,"published_at":71},"d1ea966a-493e-4abf-96d1-148c2530f7e0","美团开源LongCat-Video-Avatar 1.5：数字人视频生成从「SOTA」走向「真商用」","meituan-longcat-video-avatar-1-5-15x-dmd","2026-05-28T04:00:00+00:00",{"id":73,"title":74,"news_slug":75,"published_at":76},"ff7fa3e1-6737-4bd9-85c8-8010d13a44f3","字节跳动 Bernini 开源：用 MLLM 当\"语义规划师\"，拆开视频生成的\"思考\"与\"渲染\"","bytedance-bernini-mllm-semantic-planner-dit","2026-05-21T00:00:00+00:00",{"id":78,"title":79,"news_slug":80,"published_at":81},"f76fc8eb-1cd0-4577-aca8-db98018194ae","Mamoda2.5：DiT-MoE 架构统一多模态生成，推理速度提升 95 倍","mamoda-2-5-dit-moe-multimodal-95x","2026-05-18T13:10:00+00:00",{"id":83,"title":84,"news_slug":85,"published_at":86},"b7e375b3-6949-4ff6-9cd4-012aced6bbc3","Runway赌注视频生成通往世界模型：挑战Google的架构之赌","runway-video-world-model-google-bet","2026-05-16T04:10:00+00:00"]