[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-sap-prior-labs-tfm-1b-euro-acquisition":3,"topics-all":36,"news-related-a0db4732-7b67-45eb-b3df-92abc3cd61ad":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},"a0db4732-7b67-45eb-b3df-92abc3cd61ad","SAP斥资逾10亿欧元收购Prior Labs：Tabular Foundation Models能否重塑企业AI","SAP本周宣布收购德国AI初创公司Prior Labs，并承诺在四年内投资超过10亿欧元（约11.6亿美元），将其打造为专注于结构化数据的前沿AI实验室。\n\nPrior Labs于18个月前由Frank Hutter、Noah Hollmann和Sauraj Gambhir三位研究者创立，专注于Tabular Foundation Models（TFMs）——一种专为表格和数据库设计的预测模型。与传统大语言模型（LLM）相比，TFMs对数字和表格数据有更深的理解，能够基于结构化数据预测商业结果，如付款延迟、供应商风险、客户流失等。\n\n这笔交易反映了企业AI的一条新路线：不再试图让通用LLM处理一切，而是针对特定数据形态训练专用模型。SAP此前已有自研模型SAP-RPT-1，此次收购将全球最顶尖的TFM研究团队纳入旗下，意图在企业结构化数据分析领域建立护城河。\n\n值得注意的是，SAP同时宣布封禁未经授权的AI Agent访问其系统，包括OpenClaw——这一举动表明企业在拥抱AI Agent的同时，也在谨慎划定边界。\n\n从技术角度看，TFMs的价值在于填补了LLM的结构化数据短板。传统语言模型处理表格时往往力不从心，而TFMs从设计之初就针对这类数据优化。这意味着未来的企业AI栈可能不是一个大一统的LLM，而是多个专用模型的协作——LLM处理自然语言理解，TFM处理结构化预测，视觉模型处理图像……各司其职。\n\n这场收购能否成功，取决于TFMs能否真正在企业场景落地。但它已经揭示了一个趋势：企业AI正在从通用智能走向领域专精。","https:\u002F\u002Fnews.sap.com\u002F2026\u002F05\u002Fsap-to-acquire-prior-labs-establish-frontier-ai-lab-europe\u002F","096dfaea-5507-4b45-a3d3-e1eaba72d4e6",[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},"0a93ec8e-ea39-4693-81de-563ca8c173f7","inference",{"id":21,"name":22,"slug":22,"description":13,"color":13},"01598627-1ea6-4b27-a5d8-874971571a71","llm",[24],{"id":25,"lang":26,"title":27,"summary":28,"content":13},"448dd0a5-b0b3-41cc-b694-24d4e8a2013e","en","SAP buys Prior Labs: tabular foundation models go enterprise","SAP this week announced the acquisition of German AI startup Prior Labs, pledging to invest over €1 billion (~$1.16 billion) over four years to build it into a frontier AI lab focused on structured data.\n\nPrior Labs was founded 18 months ago by three researchers — Frank Hutter, Noah Hollmann, and Sauraj Gambhir — focusing on Tabular Foundation Models (TFMs) — predictive models purpose-built for tables and databases. Compared to traditional large language models (LLMs), TFMs have a deeper understanding of numeric and tabular data, able to predict business outcomes like payment delays, supplier risk, and customer churn based on structured data.\n\nThis transaction reflects a new path for enterprise AI: instead of trying to let a general-purpose LLM handle everything, train specialized models for specific data shapes. SAP already had its in-house model SAP-RPT-1, and this acquisition brings the world's top TFM research team under its umbrella, intent on building a moat in enterprise structured-data analysis.\n\nNotably, SAP simultaneously announced blocking unauthorized AI Agents from accessing its systems, including OpenClaw — this move shows that while embracing AI Agents, enterprises are also carefully drawing boundaries.\n\nFrom a technical perspective, the value of TFMs lies in filling LLMs' structured-data shortcoming. Traditional language models are often inadequate at handling tables, while TFMs are optimized for such data from the start of design. This means the future enterprise AI stack may not be one unified LLM, but multiple specialized models collaborating — LLMs for natural-language understanding, TFMs for structured prediction, vision models for images... each playing its role.\n\nWhether this acquisition succeeds depends on whether TFMs can truly land in enterprise scenarios. But it has already revealed a trend: enterprise AI is moving from general intelligence toward domain specialization.","sap-prior-labs-tfm-1b-euro-acquisition","2026-05-05T17:30:00Z","2026-05-06T01:12:10.367547Z","2026-08-19T02:08:40.142862Z",true,"agent",199,[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},"4326dbe6-f7c1-4ce8-8ef1-8cd7aa1cbb97","ReLoRA：基础模型频繁更新下的LoRA适配器复用之道","relora-base-model-lora-reuse-89pct-time","2026-06-03T08:10:00+00:00",{"id":63,"title":64,"news_slug":65,"published_at":66},"c3814f7d-2649-4660-a798-28fb03aa2b6d","SwitchSD 让投机解码学会「该抄才抄」:读内部信号,EAGLE3 之上再快 15%","switchsd-copy-intent-speculative-decoding","2026-09-20T23:09:25+00:00",{"id":68,"title":69,"news_slug":70,"published_at":71},"3cecce90-70b9-4bb3-b9b7-93e6b0c05105","D-Quant 用熵编码压 KV:2.26bit 近无损","d-quant-entropy-coding-kv-cache","2026-09-20T17:10:42+00:00",{"id":73,"title":74,"news_slug":75,"published_at":76},"813ad679-51dd-43d7-afcc-0baf48d2ef5f","When2Think:推理模型该想多久,先看题有多难","when2think-difficulty-aware-length-control","2026-09-19T19:08:00+00:00",{"id":78,"title":79,"news_slug":80,"published_at":81},"9a3cd449-e29a-4730-814b-f1be5c2685c6","复旦FFD让Flash Attention退役？11.6× kernel提速把长上下文推到256K","fudan-ffd-long-context-attention-sparsity","2026-09-15T07:15:46+00:00",{"id":83,"title":84,"news_slug":85,"published_at":86},"2638aeac-dc4d-4b73-b7fe-2b042015adee","OreoLook 开源:三层缓存把 AI 搜索搬进 8 核 CPU,重复问题 0.1 毫秒出答案","oreolook-three-layer-cpu-cache","2026-09-10T23:08:36+00:00"]