[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-meituan-longcat-video-avatar-1-5-15x-dmd":3,"topics-all":36,"news-related-d1ea966a-493e-4abf-96d1-148c2530f7e0":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},"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」","美团 LongCat 团队于 5 月 25 日正式开源 LongCat-Video-Avatar 1.5，一款面向商业级应用的数字人视频生成模型。作为前代开源 SOTA 模型的升级版本，1.5 在唇形同步、物理合理性、长视频稳定性、多人互动和推理效率五个维度实现了全面跃升。\n\n核心技术升级首先体现在音频编码器：从 Wav2Vec2 升级为 Whisper-large，更大的参数量和更丰富的多语言先验使模型能够更细致地捕捉音素变化与韵律特征，从而让唇形与语音的协同更精准，面部表情、头部姿态与肢体动作的时序稳定性同步提升，有效减少了长视频中的抖动与身份漂移问题。\n\n数据体系是另一大亮点。美团构建了一套多阶段数据处理流程，包括离线标注、在线验证，以及针对多人场景、沉默状态和情绪变化的专项增强数据，有效解决了开放域数字人生成中角色多样性与表达自然度的核心难题。\n\n在偏好对齐层面，1.5 引入逐帧级 GRPO（Group Relative Policy Optimization），将奖励信号细化到逐帧层面，精准修正动作不连贯、手部变形等局部缺陷。同时采用 DMD（Distribution Matching Distillation）将生成步数从 50 步压缩至 8 步，配合「基础模型 + LoRA 适配器」方案，推理效率提升约 15 倍，生成 10 秒视频仅需约 1 分钟。\n\n综合评测结果显示，LongCat-Video-Avatar 1.5 在用户偏好上优于 Kling Avatar 2.0（胜率 65.9%）、OmniHuman-1.5（61.1%）和 HeyGen（54.3%）。其跳帧问题率仅 0.8%，唇形同步问题率 29.8%，均为对比模型中最低。\n\n数字人视频生成赛道正在经历从「展示效果」到「真实部署」的转折。1.5 的开源价值不只在于模型本身，而在于将高效率生成方案开放给社区，让更多开发者可以在电商直播、在线教育、虚拟主播等场景中验证技术边界。这条从「实验室 SOTA」到「可落地的 SOTA」的路，才是开源数字人视频模型真正需要回答的问题。","https:\u002F\u002Ftech.meituan.com\u002F2026\u002F05\u002F25\u002Flongcat-video-avatar-1.5.html","76854921-bffc-4fa1-9c8f-e3269ad44d1b",[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},"b9bd9039-fcdb-41a8-b85b-fc1587def2b9","open-source",{"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},"d18679a4-e790-42bf-a1eb-1b3a0b5e3c16","en","Meituan's LongCat-Video-Avatar 1.5: from demo SOTA to real use","Meituan's tech team open-sourced LongCat-Video-Avatar 1.5, a digital human video generation model. The release focuses on \"commercial usability\" — stability, controllability, and integration with Meituan's existing products. The open-source release makes it one of the strongest digital human models available.","meituan-longcat-video-avatar-1-5-15x-dmd","2026-05-28T04:00:00Z","2026-05-28T04:05:56.847957Z","2026-08-19T02:08:40.142862Z",true,"agent",279,[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},"caed836e-2168-418f-b5c1-bde3ce962e66","Mask Forcing 往蒸馏 rollout 里掺干净 token:修视频生成的模式坍缩,指令遵循最高涨 6.5 分","mask-forcing-video-diffusion-distillation","2026-09-09T23:08:37+00:00",{"id":63,"title":64,"news_slug":65,"published_at":66},"02c8b500-ec11-44a6-8c58-6e880563dad8","FastH3 开源:4 步蒸馏版 MiniMax H3,B200 单卡最高提速 14 倍","fasth3-4-step-distilled-minimax-h3","2026-08-30T21:30:00+00:00",{"id":68,"title":69,"news_slug":70,"published_at":71},"18d2aa73-7244-4b10-b611-46475e17327e","ForgeWM开源:一步去噪72FPS的可玩世界模型,8张卡复现全流程","forgewm-few-step-playable-world-model","2026-08-24T21:10:00+00:00",{"id":73,"title":74,"news_slug":75,"published_at":76},"0599b775-ac17-49d2-aebd-a16f531c7168","腾讯混元 MeanFlowNFT：把 RL 接进「平均速度生成器」，Wan 2.1 4 步反超 50 步 LongCat-Video RL","tencent-hunyuan-meanflownft","2026-07-16T12:00:00+00:00",{"id":78,"title":79,"news_slug":80,"published_at":81},"57d2b65a-7eac-43fc-8939-ec758da2a516","Microsoft Mirage 把 3D 场景塞进扩散潜空间：视频世界模型的\"绕回去就变脸\"终于有解了","ms-mirage-3d-scene-diffusion-latent-10x","2026-06-17T10:25:00+00:00",{"id":83,"title":84,"news_slug":85,"published_at":86},"42000848-8333-40ee-ad4e-b1123dfebb0c","LoomVideo 开源 5B 统一视频生成与编辑模型：北大用「零开销」编辑机制砍掉 5.4× 推理成本","loomvideo-5b-pku-zero-overhead-5-4x","2026-06-15T21:30:00+00:00"]