当模型不再稀缺，真正稀缺的，是能把模型变成现金流的能力。 据追风交易台消息，2月9日，摩根大通证券（中国）发布研究报告《中国人工智能行业：全球布局与模型创新驱动新一代领军者》，首次覆盖中 国独立大模型厂商智谱与MiniMax。 报告开宗明义：中国人工智能行业正从"百模大战"阶段，迈向以商业化落地能力、模型创新实力及全球化布局为决定成败之关键的阶段。中国AI 市场正在迅速整合，"具备实力且资金充足的模型开发商数量已从超200家缩减至不足10家。" 摩根大通犀利指出，国内AI行业最大的利润池恐将流向掌握分发的平台巨头；而独立厂商的突围，则取决于谁能通过"结构性中立"找到生存缝隙 ——智谱向内深耕高合规的本地化部署，MiniMax向外拓展高溢价的全球市场。 这一判断的背景并不复杂。报告指出，随着模型训练成本、算力获取门槛和商业化难度持续上升，资本与算力的硬约束开始主导行业结构。换句 话说，行业已经不再奖励"能不能做模型"，而是奖励"能不能长期活下来"。 在摩根大通看来，这一阶段的核心变化在于：模型能力逐步趋同；资金消耗呈指数级上升；客户开始更关注"交付能力、稳定性与可持续性"。 | | | 市值 | | | 评级 ...

Group 1 - Hangzhou is a significant hub in the field of artificial intelligence large models, with Alibaba's large model performing exceptionally well [1] - Hangzhou Jinx Technology Co., Ltd. focuses on developing a general brain engine for robots to address the compatibility issues between humanoid robot software and hardware [1] - Alibaba's real estate division signed a total leasing area of 500,000 square meters nationwide for the fiscal year 2025, with 410,000 square meters located in Hangzhou [1] Group 2 - The Alibaba Digital Ecological Innovation Park hosts over 370 enterprises, primarily focusing on the artificial intelligence industry, with different parks emphasizing various aspects [1] - Zhejiang Shiyue Technology Co., Ltd. is another company located in Alibaba's park, specializing in dexterous hand operation models and data acquisition solutions [1] - The CEO of Shiyue Technology stated that the company aims to leverage Alibaba's platform to connect with more upstream and downstream enterprises, including humanoid robot companies and suppliers of key modules [2] Group 3 - Shiyue Technology plans to officially launch the mass production version of its dexterous hand in February, with an expected price of around 100,000 yuan, to be listed on JD's artificial intelligence platform [3]

来源：环球网 【环球网科技综合报道】2月10日消息，据9to5mac报道，一位匿名知情人士透露，OpenAI已经正式启 动ChatGPT内部广告测试，成为生成式AI领域首个尝试商业化广告的头部企业。 OpenAI强调，此次测试为"初步探索阶段"，部分实现细节（如广告主筛选、频次控制）或在未来数周 至数月内调整，但"广告作为产品生态一部分"的定位不会改变。 尽管广告被视为OpenAI规模化盈利的重要路径，但知情人士称，该公司预计长期广告收入占比将低于 总营收的50%，其余收入可能来自企业级API服务、订阅制（如ChatGPT Plus）及定制化AI解决方案。 值得关注的是，就在OpenAI启动测试前一日（超级碗赛事期间），其竞争对手Anthropic投放了一则引 发争议的广告，画面中仅显示文字"人工智能领域禁止广告"（No Ads in AI），被广泛解读为针对 OpenAI的直接回应。（青山） 此前，该公司曾公开承诺广告将"清晰标识、仅置于回答底部且不影响内容生成"，此次测试被视为其探 索AI产品盈利模式的关键一步。 根据OpenAI上月披露的方案，广告将仅出现在ChatGPT聊天窗口的回答区域底部，以非侵入 ...

分析指出，GLM系列模型近年来在代码生成和智能体能力上的进步有目共睹，这与Pony Alpha的主打方 向完全一致；智谱首席科学家唐杰发微博明确表示最近智谱将发布GLM-5。目前，智谱等疑似关联方 尚未对Pony Alpha的身份作出官方回应。 消息面上，2月6日，全球模型服务平台OpenRouter悄然上线一款代号为"Pony Alpha"的匿名模型，因其 强大的编码能力、超长上下文窗口及针对智能体工作流的深度优化，迅速引发开发者社区关注。市场猜 测该模型的身份可能是智谱即将发布的新一代模型GLM-5。 智通财经APP获悉，智谱(02513)午前涨超22%，高见338.2港元，较招股价116.2港元已涨近200%，总市 值逼近1500亿港元。截至发稿，涨21.17%，报335.4港元，成交额6.65亿港元。 ...

机器之心编辑部 2026 年刚拉开序幕，大模型（LLM）领域的研究者们似乎达成了一种默契。 当你翻开最近 arXiv 上最受关注的几篇论文，会发现一个高频出现的词汇： Self-Distillation 。 近年来，基础模型取得了显著的成功，为语言、视觉、机器人等领域的 AI 应用提供了强大的支持。 但在真正落地、长期使用的过程中，研究者逐渐发现：如何让模型在不断吸收新知识的同时，不丢失已有的核心能力 —— 即「持续学习」，正成为制约大 模型进化的关键瓶颈。 传统的强教师依赖范式因成本与数据依赖，难以适配高频的持续进化。 Self-Distillation（自蒸馏） 随之成为破局点 —— 通过合理的上下文引导或反馈机 制 ，模型完全可以构建出一个比当前权重更聪明的临时自我，让模型在没有外部强教师的情况下实现内生增长。 基于这一深刻洞察，由 MIT、ETH Zurich、Meta 及斯坦福等顶尖机构组成的紧密学术圈，在 2026 年 1 月密集发布了三项研究成果。 1.Self-Distillation Enables Continual Learning 在持续学习领域，传统的监督微调（SFT）常因 「灾难性 ...

价值锚点："今天排"智能体，解耦复杂排产 目前，圆木智能从"排产"这一具体场景切入，为应对"多品种小批量、订单变化频繁、工序复杂设备 多"等企业排产中面临的复杂挑战，推出了"今天排"智能体产品。"今天排"通过大模型理解生产约束， 融合优化算法实现动态排程，将排产时间从小时级缩短至分钟级，面对插单等异常情况，系统可快速生 成新方案。同时，"今天排"能适配多源数据输入，降低使用门槛，实现产品在排产场景中规模化落地。 目前，产品已在苏州、无锡等地机加工工厂投产使用，帮助客户实现全工序自动排程，排产时间缩短 90%，订单交付率提升15%以上，有效提升设备稼动率与交期确定性。 圆木智能核心成员均毕业于清华大学、北京大学等顶尖学府，兼具人工智能前沿研发、成功SaaS产品从 0到1商业化、以及超过二十年的工业数字化实战与销售经验。"算法+工业+产品"的复合背景，使团队 既能够洞察工业领域复杂性背后的核心痛点，又具备将新一代AI技术转化为稳定、易用产品的研发能 力。此外，圆木智能已与百度、智谱等大模型厂商达成深度合作，与清华、浙大、复旦等高校实验室建 立产学研合作关系，不断巩固技术优势，共同夯实工业智能的基座。 未来蓝图：从 ...

Group 1 - The core focus of the article is on the emphasis of innovation in the development strategy of the Shangcheng District, highlighting the importance of the Central Innovation District (CID) as a key component of the district's goals during the 14th Five-Year Plan [2][6][9] - The presence of over ten AI companies showcasing their latest technologies at the district's two sessions reflects the active engagement in innovation and the practical implementation of the CID strategy [2][3][5] - The district's GDP reached 2900.5 billion yuan during the 14th Five-Year Plan, with significant contributions from the financial sector, indicating a strong economic foundation but also revealing challenges in enhancing innovation capabilities [6][9] Group 2 - The article discusses the establishment of a concentrated display area for AI companies at the two sessions, allowing representatives to experience the innovation firsthand, which includes products like AI glasses and high-speed cameras [3][4][5] - The CID has rapidly progressed from conceptualization to practical implementation, with the establishment of various innovation hubs and the introduction of 45 core AI enterprises, indicating a vibrant ecosystem for technological advancement [8][9] - The article outlines the need for a robust talent ecosystem and the importance of nurturing young talent to sustain innovation momentum, suggesting initiatives like the "CID Youth Talent Special Plan" to support this goal [10][11]

宗建树 长江证券分析师: 各位领导，大家晚上好，我是长江证券的首席分析师宗建树，今天由我给大家汇报一下我 们整个的一个 AI 势不可挡，2026 年模型升级有哪些预期差的一个整体的汇报。然后 AI 是围绕是我们最近开的一个新的系列，因为最近整个 AI 的产业其实近有一波比较大的一 个调整。但是我们觉得最近调整的一个主要核心原因，第一个就是确实现在在整个需求侧 落地，目前还没有看到明显的加速。第二个是海外的宏观的一个波动，也进步放大了整个 AI 的波动。 但是，我们认为从整个 AI 大的产业趋势来看，目前的产业趋势的确定性是在不断的提升， 所以我们坚定看好整个后续。产业趋势，整个后续的一个发展。今天主要汇报的是关于模 型方面的升级，因为从这个 AI 来看，模型是这个 AI 最大的核心的一个驱动力。我们觉得 在整个 2026 年，我觉得是一个模型升级原有的范式的曲线继续向上，然后模型又逐渐开 始跟场景融合的一个年份。所以今年第一个模型的演进，一样会持续的一个向上。第二个 我们觉得整个模型的，跟场景结合之后，它的落地会全面的加速。 所以，这是我们非常去看好整个 2025 年整个 AI 去表现的一个非常重要的原因。我 ...

Core Insights - The report by Qingcheng Jizhi and Huqing Puzhi AI Incubator analyzes the application of large model API services in content creation, code development, and professional services, highlighting their impact on daily work routines and productivity [1][3]. Group 1: Code Development - Developers face significant time consumption in tasks such as code completion, bug debugging, and multi-file understanding, which exhibit "short input, medium output" characteristics, posing challenges for model context stability and response speed [1][3]. - GLM and DeepSeek series model APIs are becoming the preferred efficiency tools for developers due to their coding capabilities and long context advantages [1][5]. - API usage shows a unique "nighttime double peak" distribution, with high activity between 21-23 and 1-2 AM, indicating programmers' focused work hours [5]. Group 2: Content Creation and Marketing - Large models have become essential tools for content creation, assisting in rapid generation of copy and proposals, as well as in content marketing through expansion and stylization [5]. - Kimi and MiniMax series models are particularly favored in these scenarios, significantly reducing repetitive creative tasks and enhancing the novelty of marketing content [5]. Group 3: Professional Services and Office Automation - In professional services, such as legal and financial document processing, the focus is on stability and speed, with tasks often involving short to medium input and medium output interactions [2][5]. - Qwen and MiniMax series models are preferred for automating office processes, improving efficiency and accuracy in high-frequency, low-creative tasks like contract review and data analysis [2][5]. - The report emphasizes that individual success is foundational to corporate success, with enhanced personal efficiency driving overall business performance [6].

Group 1 - The core viewpoint of the news highlights the strong performance of AI concept stocks in the Hong Kong market, particularly the company Zhiyu (2513.HK), which saw its stock price surge by over 20% after a previous increase of 36%, reaching a new high of 333 HKD and a market capitalization nearing 150 billion HKD [1]. - Zhiyu has been active in the large model sector, with its hybrid thinking model GLM-4.7-Flash achieving over 1 million downloads on Hugging Face within two weeks of its release [2]. - Another model, Pony Alpha, recently launched on the global model service platform OpenRouter, has gained attention for its strong coding capabilities and deep optimization for intelligent agent workflows, showcasing significant advantages in AI applications [3]. Group 2 - The stock performance of Zhiyu reflects a broader trend in the AI sector, indicating investor confidence and interest in companies involved in advanced AI technologies [1]. - The success of the GLM-4.7-Flash model and its rapid adoption suggests a growing demand for innovative AI solutions in the market [2]. - Pony Alpha's capabilities, including high accuracy in tool invocation and impressive performance in programming tasks, highlight the competitive landscape and potential for further advancements in AI models [3].