[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-sensetime-sensenova-u1-pro":3,"news-related-7c5d7da6-2bdb-4174-b28b-d7fef5579181":36},{"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},"7c5d7da6-2bdb-4174-b28b-d7fef5579181","商汤 SenseNova U1 Pro 把多模态 AI 卷出「长程交付」:从「好看」走向「可用」的赛道切换","商汤在 WAIC 2026 现场正式发布 SenseNova 系列旗舰 SenseNova U1 Pro。与多数厂商持续堆「生成更像」指标不同,U1 Pro 把核心定位放在「长程任务的交付级原生多模态智能体基座」,瞄准的是把多模态从「灵感素材工具」升级成「系统级任务交付单元」。\\n\\nU1 Pro 在 NEO-unify 内核之上构建四项核心交付能力:专业级设计质感、原生 8K 超清输出、高密度图文细节控制、长程智能体闭环思维。底层是统一基座内生的「理解、生成、行动」能力,让模型可以围绕复杂目标执行数十轮智能体生成循环,边改边续。\\n\\n实战侧,U1 Pro 一次性输出 4:1 宣纸水墨长卷《智会世图》(WAIC 九年纪事)、《沙影之刃》22 个连贯分镜的世界观设定,并在世界杯开赛前一个月预测到黑马佛得角出线 —— 把「信息搜集 → 分析推理 → 复杂视觉交付」做成完整闭环。预览版即日起邀测,正式版 2026 年 8 月对外开放并同步上线 API。\\n\\n从行业看,真正的赛点已不在图像生成质量,而在「能否稳定交付专业级成果」。U1 Pro 这条路走通,意味着国产多模态开始把「可用」放在「好看」之前 —— 这才是 AGI 走进真实生产链路的关键一步。","https:\u002F\u002Fnews.qq.com\u002Frain\u002Fa\u002F20260718A0AIKD00","d46ec0a7-501b-4ef8-9c89-2391b2701b3b",[10,14,17,20],{"id":11,"name":12,"slug":12,"description":13,"color":13},"a8002d98-9df1-4ab9-94d4-a7625af634c4","china-ai",null,{"id":15,"name":16,"slug":16,"description":13,"color":13},"01598627-1ea6-4b27-a5d8-874971571a71","llm",{"id":18,"name":19,"slug":19,"description":13,"color":13},"7e89b5cc-57db-4f37-bc6d-28919a73931c","model-release",{"id":21,"name":22,"slug":22,"description":13,"color":13},"499f4b56-819d-49a3-9609-33e775143b86","multimodal",[24],{"id":25,"lang":26,"title":27,"summary":28,"content":28},"0916830f-bd42-43bc-b17e-0e0a04ad908c","en","SenseNova U1 Pro: multimodal AI from demo to delivery","SenseTime officially launched its SenseNova U1 Pro flagship at WAIC 2026. Unlike most vendors continuing to pile on \"more like\" generation metrics, U1 Pro anchors on \"long-horizon task delivery-grade native multimodal agent foundation\" — upgrading multimodal from \"inspiration-material tool\" to \"system-level task delivery unit\". Built on the NEO-unify core, U1 Pro offers four delivery capabilities: professional-grade design aesthetic, native 8K ultra-clear output, high-density text-image detail control, and long-horizon agent closed-loop thinking. Underneath, the unified base's intrinsic \"understand, generate, act\" capabilities let the model run dozens of agent-generation loops around complex goals, iterating and extending on the fly. On the practical side, U1 Pro outputs in one shot a 4:1 rice-paper long-scroll \"智会世图\" (a WAIC ninth-anniversary chronicle), the 22-frame worldview setting of \"Sand Shadow Blade\" (沙影之刃), and predicts the dark-horse Cape Verde qualifying for the World Cup a full month before the tournament — turning \"information gathering → analytical reasoning → complex visual delivery\" into a closed loop. The preview is now in invite-only testing; the official version will open in August 2026 and ship with an API at the same time. Industry-wise, the real race is no longer image-generation quality but \"stable delivery of professional-grade output\". U1 Pro's path means domestic multimodal is starting to put \"usable\" ahead of \"pretty\" — the real step for AGI to enter real production.","sensetime-sensenova-u1-pro","2026-07-19T10:01:00Z","2026-07-19T18:10:11.320096Z","2026-08-19T02:08:40.142862Z",true,"agent",96,{"items":37},[38,43,48,53,58,63],{"id":39,"title":40,"news_slug":41,"published_at":42},"a151db0c-d832-4df2-ac03-2d4e58b26e99","Kimi K3 跑通 MiniTriton:Moonshot 让 LLM 第一次从零编译出自己的 GPU 编译器","kimi-k3-minitriton-gpu-compiler","2026-07-26T14:00:00+00:00",{"id":44,"title":45,"news_slug":46,"published_at":47},"cc43635e-9f4c-4007-b9e9-347b02f67a76","文远知行 WITT:用\"原子级物理事实\"重写自动驾驶的数据飞轮","weride-witt-atomic-physics","2026-07-17T08:00:00+00:00",{"id":49,"title":50,"news_slug":51,"published_at":52},"f22a15d1-07a0-4318-a5de-6cc794a54d8b","全国首个具身世界模型「我悟」拿下备案：北京人形用 Pelican-VL × WoW 打开 API 商业化通道","beijing-humanoid-wow-pelican-vl-embodied","2026-06-27T10:00:00+00:00",{"id":54,"title":55,"news_slug":56,"published_at":57},"caa54bff-d57a-411c-9dae-43f1d4d46875","DeepSeek V4 重磅登场：长期记忆技术突破重塑AI能力边界","deepseek-v4-engram-ltm-long-term-memory-87pct","2026-04-22T07:05:00+00:00",{"id":59,"title":60,"news_slug":61,"published_at":62},"f6e4aab0-7693-4c2c-bb66-c1641fc2cc3e","Ox Alpha 谜底揭晓:智谱 GLM-5.3-Flash,MIT 开源 320B MoE","ox-alpha-glm-5-3-flash-reveal","2026-08-27T13:30:00+00:00",{"id":64,"title":65,"news_slug":66,"published_at":67},"804ab59a-a8d6-4b61-bf74-8f6f2bdae83c","智谱把 Flash 做成一件正经事:一次说清 GLM-5.3-Flash 的架构和 benchmark 真相","glm-5-3-flash-hybrid-attention-architecture","2026-08-27T08:00:00+00:00"]