[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-unisound-u2-266b-10b-native-agent":3,"topics-all":36,"news-related-ea397827-ba45-4c2f-9a9a-6f4871b851cb":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},"ea397827-ba45-4c2f-9a9a-6f4871b851cb","云知声 U2 大模型正式发布：266B 总参数 \u002F 10B 激活的原生 Agent 架构","云知声 6 月 8 日发布新一代通用大模型 U2，把\"原生 Agent 大模型\"标签贴到自己身上。U2 总参数 266B，单次推理仅激活约 10B——相比动辄上万亿的稠密模型，token 消耗量约为其 25%，\"智能密度\"在这组数字里被具象化了。\n\n技术路径上，U2 走了一条与多数同行不同的路线：把工具调用、状态管理、任务规划从训练第一天就内化到模型里，而非\"先练聊天模型再外挂 Agent 框架\"。为此云知声引入了**模型 + Harness 协同演进**机制——训练中模型主体越复杂，驾驭脚手架的支撑节点和验证精度也同步延伸；更严苛的脚手架反过来约束模型输出，形成闭环。\n\n训练侧还有两个关键设计：**过程监督 + 课程学习**组合，让模型从易到难、从短上下文到长链路任务渐进进化；**隐式思考推理 + 显式思考验证**——日常探索留在隐空间，避开显式 reasoning 的 token 浪费，关键阶段再切到可读、可校验的显式推理。这和 o 系列、R1 那种\"全程展开思考链\"截然相反。\n\n跑分表现也站得住。GPQA Diamond 87.9，超过 GLM-5.1（86.2）、DeepSeek-V4-Flash High（87.4）、MiniMax M2.5\u002FM2.7（87.4）；IFBench 指令遵循 77.3；AA-LCR 长上下文 70，反超 GLM-5.1（62.3）、DeepSeek-V4-Flash（62.7）。Claw-Eval pass@3 拿到 76.9，Agent 能力居国产第一梯队前列。\n\n云知声把\"高智能密度 × 高 Token 价值\"做成产品公式，本质是在回答：当算力不再无限时，企业真正买的是\"每一美元 token 能换回多少可靠任务完成度\"。U2 的 266B\u002F10B MoE 选型、Harness 协同、隐式\u002F显式推理切换，把这条路径工程化了。\"少参数也能跑出可执行 Agent\"正在悄悄改写国产基础模型的价值评估标准。","https:\u002F\u002F36kr.com\u002Fp\u002F3844393508047108","5e4fd3d1-9cb4-44a6-bae5-9ffb449c05c1",[10,14,17,20],{"id":11,"name":12,"slug":12,"description":13,"color":13},"6ad31a14-c0da-42df-81fd-564281f768db","agentic-ai",null,{"id":15,"name":16,"slug":16,"description":13,"color":13},"a8002d98-9df1-4ab9-94d4-a7625af634c4","china-ai",{"id":18,"name":19,"slug":19,"description":13,"color":13},"01598627-1ea6-4b27-a5d8-874971571a71","llm",{"id":21,"name":22,"slug":22,"description":13,"color":13},"7e89b5cc-57db-4f37-bc6d-28919a73931c","model-release",[24],{"id":25,"lang":26,"title":27,"summary":28,"content":13},"0531a8e6-bfea-427f-bc82-d9a28ee4fedf","en","Unisound U2: a 266B\u002F10B native agent architecture","Unisound released the next-generation general large model U2 on June 8, slapping the \"native Agent large model\" label on itself. U2 has a total of 266B parameters, activating only about 10B per inference — compared to dense models that easily exceed a trillion, the token consumption is about 25% of that, and \"intelligence density\" is embodied in this set of numbers.\n\nOn the technical path, U2 takes a route different from most peers: internalizing tool calling, state management, and task planning into the model from training day one, rather than \"training a chat model first and then externalizing an Agent framework.\" For this, Unisound introduces a **model + Harness co-evolution** mechanism — the more complex the model body becomes in training, the more the support nodes and validation precision of the controlling scaffold also extend in sync; the more rigorous the scaffold, the more it constrains the model output in return, forming a closed loop.\n\nThere are also two key training-side designs: a **process supervision + curriculum learning** combination lets the model evolve gradually from easy to hard, from short-context to long-chain tasks; **implicit thinking reasoning + explicit thinking verification** — daily exploration stays in latent space, avoiding the token waste of explicit reasoning, and switches to readable, verifiable explicit reasoning at key stages. This is the opposite of the \"full-chain-of-thought\" path of the o-series and R1.\n\nThe benchmark performance also holds up. GPQA Diamond 87.9, beating GLM-5.1 (86.2), DeepSeek-V4-Flash High (87.4), and MiniMax M2.5\u002FM2.7 (87.4); IFBench instruction following 77.3; AA-LCR long context 70, beating GLM-5.1 (62.3) and DeepSeek-V4-Flash (62.7). Claw-Eval pass@3 takes 76.9, putting Agent capability at the front of the domestic first tier.\n\nUnisound is turning \"high intelligence density × high token value\" into a product formula, essentially answering: when compute is no longer infinite, what enterprises really buy is \"how much reliable task completion per dollar of token.\" U2's 266B\u002F10B MoE choice, Harness co-evolution, and implicit\u002Fexplicit reasoning switching engineer this path. \"Few parameters can still run a deliverable Agent\" is quietly rewriting the value-evaluation standard for domestic foundation models.","unisound-u2-266b-10b-native-agent","2026-06-09T00:00:00Z","2026-06-09T00:28:16.278475Z","2026-08-19T02:08:40.142862Z",true,"agent",197,[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},"d1e8997e-bb60-453d-9ef8-71b8bdde5386","Harvey 首个自研法律模型 Tenet 曝光:底座没选 GPT 和 Claude,选了 Kimi K3","harvey-tenet-kimi-k3-legal-model","2026-08-18T17:30:00+00:00",{"id":63,"title":64,"news_slug":65,"published_at":66},"829523ee-978e-45ca-8692-62d9217864cc","Qwen-AgentWorld：千问把语言世界模型做成 Agent 的统一入口","qwen-agentworld-language-world-model","2026-06-24T06:05:00+00:00",{"id":68,"title":69,"news_slug":70,"published_at":71},"8ebbcd9c-31ee-4baa-b395-b104bd87c8e1","Kimi K2.8 Preview 把 K3 的百万上下文下放给免费档：月之暗面的「过日子」模型登场","kimi-k2-8-preview-coding","2026-09-17T03:00:00+00:00",{"id":73,"title":74,"news_slug":75,"published_at":76},"2b37a19b-1dde-4238-bef5-39b1d19157f1","OpenBMB 开源 MiniCPM5-2B:2B 端侧模型平均分超对比集 4B 级","openbmb-minicpm5-2b-on-device","2026-09-07T17:02:00+00:00",{"id":78,"title":79,"news_slug":80,"published_at":81},"d400c0db-49df-4cc6-a87e-87b709f59fea","Muse Spark 1.3 发布:卡住会向用户求助的 Agent,工具调用少 20%、token 省 25%","muse-spark-1-3-meta-agent-release","2026-09-06T15:12:00+00:00",{"id":83,"title":84,"news_slug":85,"published_at":86},"453ce9a1-5d55-4981-b44d-c261b8051724","GLM-5.3 753B 权重上架 HuggingFace,智谱兑现两周开源承诺","glm-5-3-weights-huggingface-release","2026-08-28T15:15:00+00:00"]