[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-minicpm-samsung-end-side":3,"topics-all":36,"news-related-63506aeb-566f-47dc-a59f-d4f0197e4545":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},"63506aeb-566f-47dc-a59f-d4f0197e4545","面壁 MiniCPM 拿下三星旗舰端:7 款端侧 LLM 同日备案,标志端侧 AI 进入合规量产节奏","7 月 15 日,中国端侧大模型迎来少见的「集结点验」:网信办同日公布 7 款手机端侧生成式 AI 完成备案,涵盖 Apple 智能、华为小艺、OPPO Andes GPT、vivo 蓝心、小米澎湃、三星盖乐世 AI 与努比亚豆包手机大模型。同时,36 氪独家披露,面壁智能 MiniCPM 系列将搭载三星旗舰。MiniCPM 的价值在于 1B-4B 参数区间把量化、稀疏和指令压缩做到手机 SoC 上能稳定跑:MiniCPM5-1B 以 17.9 智能指数登顶 1B 级效率榜,MiniCPM-V 4.6 用 1.3B 多模态覆盖三大手机系统,4-bit 量化仍保九成能力。这种「小而稳」是手机厂商敢把它写进旗舰 ROM 的前提。更深层信号在产业分工。三星此前以 Galaxy AI 加 Gemini 为云端兜底,面壁进入供应链意味着「终端 + 模型」开始分层:终端厂商负责硬件、系统集成与合规,模型公司把能力压进 SoC 内存与功耗预算。云端拼综合能力与 token 单价,端侧拼的则是 12GB 内存、3 瓦功耗、严格隐私下的稳定交付。「搭载」不等于「主导」也不等于规模化收入;具体机型、入口与推理路径仍待验证。但 7 款备案同日通过本身即标志拐点:端侧 AI 从 demo 阶段跨入合规量产阶段。","https:\u002F\u002F36kr.com\u002Fp\u002F3896830362601351","5e4fd3d1-9cb4-44a6-bae5-9ffb449c05c1",[10,14,17,20],{"id":11,"name":12,"slug":12,"description":13,"color":13},"5e628969-6d2a-437f-998a-104e4b16cfb1","ai-progress",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},"499f4b56-819d-49a3-9609-33e775143b86","multimodal",[24],{"id":25,"lang":26,"title":27,"summary":28,"content":13},"d867f214-eb31-4487-921b-88d38480d3ce","en","MiniCPM lands on Samsung flagships; 7 on-device LLMs filed","On July 15, China's on-device LLMs hit a rare \"rallying point\": the Cyberspace Administration of China announced on the same day that 7 on-phone on-device generative AIs had completed filing — covering Apple Intelligence, Huawei's Xiaoyi, OPPO's Andes GPT, vivo's BlueLM, Xiaomi's Pengpai, Samsung Galaxy AI, and Nubia's Doubao Phone LLM. Meanwhile, 36Kr exclusively revealed that ModelBest's MiniCPM series will be featured on Samsung's flagship. MiniCPM's value lies in getting quantization, sparsity, and instruction compression to run stably on a phone SoC within the 1B-4B parameter range: MiniCPM5-1B tops the 1B efficiency ranking with a 17.9 intelligence index, MiniCPM-V 4.6 uses 1.3B multimodality to cover the three major phone systems, and 4-bit quantization still retains 90% of capability. This \"small but stable\" is the prerequisite for phone vendors to write it into their flagship ROM. A deeper signal lies in the industry division of labor. Samsung previously used Galaxy AI + Gemini as the cloud-side fallback; ModelBest entering the supply chain means \"device + model\" is starting to be layered: device vendors handle hardware, system integration, and compliance, while model companies squeeze capabilities into SoC memory and power budgets. The cloud side competes on overall capability and per-token pricing; the device side competes on stable delivery under 12GB memory, 3-watt power, and strict privacy. \"Being included\" doesn't equal \"leading\" or scaled revenue; specific models, entry points, and inference paths are still to be verified. But 7 filings passing the same day itself marks an inflection point: on-device AI has crossed from the demo stage to the compliant mass-production stage.","minicpm-samsung-end-side","2026-07-15T14:01:00Z","2026-07-15T14:08:04.724294Z","2026-08-19T02:08:40.142862Z",true,"agent",125,[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},"bd050dd6-4d85-4616-a004-55c23c533a24","腾讯混元合并大语言模型与多模态团队，成立基础模型部探索全模态统一","tencent-hunyuan-foundation-model-dept","2026-07-24T03:00:00+00:00",{"id":63,"title":64,"news_slug":65,"published_at":66},"6b349d2c-3d03-4cb9-8f47-63e8c288d0db","美方三机构联合指控六家中国 AI 企业系统性蒸馏美国模型","us-accuses-six-chinese-ai-firms-of-distillation","2026-09-10T01:08:40+00:00",{"id":68,"title":69,"news_slug":70,"published_at":71},"0d8fdf45-4585-47c0-9e78-3652e318b156","Apple Intelligence 中国版落地:通义千问接管语言 AI,百度负责视觉搜索","apple-intelligence-china-qwen-baidu-2026","2026-08-25T12:00:00+00:00",{"id":73,"title":74,"news_slug":75,"published_at":76},"1844afb1-3a1c-4acd-9e4c-f5e2792a2018","下载免费不等于商用免费：HF Summer 2026 隐藏的开源前沿许可证分水岭","frontier-license-shift-hf-summer-2026","2026-08-23T12:30:00+00:00",{"id":78,"title":79,"news_slug":80,"published_at":81},"2fc64783-8b2a-49a3-939b-edf02bff3622","Ox Alpha 指纹指向 GLM-5.3:OpenRouter 的 1M 上下文隐身模型可能是智谱","ox-alpha-glm-5-3-stealth-zhipu","2026-08-22T14:00:00+00:00",{"id":83,"title":84,"news_slug":85,"published_at":86},"9389d1ed-dd2d-41cb-bbc5-9a543e2b2f71","开源报告里的「参数天花板」分水岭:中国实验室把上限拉到2.78T,美国还在130B徘徊","hf-summer-2026-china-open-weight-parameter-ceiling","2026-08-20T06:00:00+00:00"]