大家好，我是歸藏（guizang），今天给大家带来昨天探索的 AI 许愿祈福壁纸教程。 我也录了一个很详细的视频教程，如果你懒得看文字可以直接看视频，方便的话也可以帮我点个赞，谢谢。 昨天做了几张 AI 玄学的那种祈福壁纸，除了常见的文字花纹还加上了对应的神仙和一些现代化的处理。 另外还搞了一个手机的动态壁纸，解锁就会播放视频非常的炫。 这篇内容就教一下大家我是如何写提示词和发散创意的，主要是后面的部分学会的话你可以从任何小的点发散 属于自己的创意。 先发提示词模板，后面是创作思路： 主体为以复古票据为原型，米黄底色，外围有繁复绿纹边框 钟馗，手持电话，闪光灯拍照，现代潮流服饰，板鞋，艳丽色彩，荧光油绘，夸张的线条，夸张的 姿态，梦幻光影，水墨电影感。 中央用粗犷黑笔写着 '小人退散'，顶部和底部有重复英文 'EVERYTHING GOES WELL'，中间 'GO OD LUCK'，两侧竖排英文 'LIFE IS SHORT WHY NOT TRY'，周遭簇拥竖排小字，底部有英文 'Wi sh you all the heavenly blessings' 与红色篆刻印章的画面，国潮票据场景，复古花纹、书法 ...

我去！ Black Forest Labs 和 Krea 一起开源了一个新的图像模型 FLUX.1-Krea [dev] 专注于打造具有独特美感的图像。没有"AI 效果"，没有过曝的高光，只有自然的细节。 而且这个模型完全可以兼容之前的 FLUX 开源模型生态系统，这个太重要了。 而且他们发布了一个技术报告，详细介绍了模型的实现思路和训练过程，也介绍了一下 AI 感出现的原因，这 部分更重要，我总结和分析一下。 先看案例 解析"AI 风格" "当一个指标成为目标时，它就不再是一个好的指标" —Charles Goodhart 大家最近对 AI 脸和 AI 质感诟病都很多，在使用 AI 生成图像时，一个明显的趋势是它们独特的外观：过于模 糊的背景、蜡质的皮肤质感、乏味的构图等等。这些问题共同构成了现在所谓的"AI 风格"。 人们常常关注模型有多"聪明"。我们经常看到用户测试复杂的提示词。它能让马骑上宇航员吗？它能把酒杯 倒满吗？它能正确渲染文字吗？ 多年来，我们设计了各种基准来将这些问题形式化为具体的指标。研究界在推动生成模型方面取得了显著成 就。 然而，在追求技术能力和基准优化的过程中，早期图像模型中那种杂乱 ...

Core Viewpoint - Trickle AI is revolutionizing the Vibe Coding ecosystem by providing a more efficient and user-friendly platform for web development, significantly reducing the time and cost involved in creating and modifying web pages [2][12][67]. Group 1: Introduction to Trickle AI - Trickle AI offers an advanced interface that changes the way Vibe Coding is approached, necessitating new principles for interaction with AI agents [2][12]. - The platform allows users to build complete products efficiently, addressing previous limitations faced with other coding agents [12][67]. Group 2: Features and Capabilities - The Magic Canvas feature provides a permanent context for web development, allowing users to manage databases, assets, and knowledge effectively [19][67]. - Users can modify projects quickly and cost-effectively using the Edit mode, which simplifies the process of making style and content changes [21][24][67]. - Trickle AI integrates design variables, enabling users to make consistent style changes across multiple pages without excessive token consumption [29][31][35]. Group 3: Database Integration and Functionality - Trickle AI allows for easy database integration, enabling users to standardize and upload data efficiently [36][40]. - The platform supports the creation of backend functionalities to manage data uploads and synchronization with external services like Algolia for search capabilities [53][56]. Group 4: Website Optimization and Launch - Trickle AI provides tools for SEO optimization, custom domain binding, and data analysis, essential for effective website management post-launch [59][60][66]. - Users can enhance the aesthetic appeal of their websites through various design modifications and the addition of interactive components [43][47][51]. Group 5: Future Implications and Recommendations - The evolution of Trickle AI signifies a shift in web development paradigms, moving towards a more integrated and user-centric approach [71][72]. - Developers are encouraged to focus on system thinking, leveraging AI as a cognitive tool rather than a mere replacement, and to establish a collaborative relationship with AI [72].

Core Insights - The article discusses the release of GLM-4.5 by Zhipu, highlighting its strong performance in reasoning, coding, and agent capabilities, with a total parameter count of 335 billion and an activation parameter count of 32 billion [1] - GLM-4.5 is noted for its cost-effectiveness, priced at 0.8 yuan per million tokens for input and 2 yuan per million tokens for output, with a high-speed output rate exceeding 100 tokens per second [1] Performance and Features - GLM-4.5 demonstrates superior coding abilities, even with fewer total parameters compared to competitors, and excels in mixed reasoning, providing excellent results even with short prompts [2] - The model integrates various agent capabilities within a single API, allowing for seamless product development and the creation of a simplified ChatGPT-like agent [3][25] - It is compatible with Claude Code, enabling users to replace Claude Code models easily [5] Use Cases and Applications - The model successfully completes coding tasks without complex instructions, such as generating a Gmail page or a 3D abstract art piece, showcasing its ability to understand and execute detailed requirements [7][9] - GLM-4.5 can create comprehensive components like a calendar manager and an OKR management tool, fulfilling all specified requirements without bugs [11][13][14] - The model also generates high-fidelity e-commerce web pages, including detailed checkout processes, demonstrating its capability in UI/UX design [17][19][20] Integration and Accessibility - GLM-4.5 supports integration with various tools and APIs, including a search tool for generating dynamic web pages based on real-time data, such as event information for WAIC [27][28] - The model is available for a subscription fee of 50 yuan for unlimited usage, making it accessible for developers and non-developers alike [34] Strategic Positioning - The article emphasizes that GLM-4.5 represents a strategic advantage by integrating multiple functionalities into a single model, contrasting with competitors that have developed fragmented solutions [35][36] - This integration approach allows users to streamline their workflows, reducing the need for multiple models and simplifying the process of cross-model orchestration [36][37]

Core Viewpoint - The article introduces Myshell ShellAgent 2.0, highlighting its ability to create agent applications with minimal input, significantly lowering the barrier for users to develop interactive tools without complex front-end requirements [1][2][21]. Summary by Sections Agent Creation Process - The creation process for agents is simplified, requiring only a user’s needs to be articulated without concern for interface design [2][21]. - Users can generate agents by providing a single prompt, which initiates a demand analysis and prompts for additional details before generating the agent [4][21]. Case Study: Fortune Telling Agent - An example is provided where a user requests an agent for fortune-telling based on birth date, showcasing the ease of input and the professional output generated [3][7]. - The agent performs a comprehensive analysis, including a detailed breakdown of the user's fortune, personality traits, and suggestions for career and financial management [8]. Web3 Wallet Analysis Tool - Another application discussed is a tool that analyzes Web3 wallets, allowing users to understand asset movements and transactions in an entertaining format [13][15]. - The tool aims to make complex blockchain data accessible and engaging for users unfamiliar with the technology [13][15]. Learning Tools - The article also describes a feature that converts lengthy documents or articles into interactive flashcards or audio summaries, enhancing the learning experience [17][18]. - Users can upload documents or links, and the system will summarize key points and generate study materials [17][20]. User Engagement and Accessibility - The platform encourages user engagement by allowing anyone to create their own agents with simple ideas, thus democratizing access to technology [21][22]. - The article concludes with an invitation for readers to explore the platform and participate in a giveaway, promoting community involvement [22][23].

Core Insights - Lovart has introduced a significant update that enhances user interaction with its design agent, transforming the user experience from a tool-centric to an agent-centric model [1][29][33] Group 1: Update Features - The new update includes a comprehensive commenting system called ChatCanvas, allowing users to interact directly with the design agent [3][4] - Users can now provide specific feedback on images by clicking on them and writing comments, making the design process more intuitive [11][20] - The agent can understand and complete user requests through a predictive text feature, enhancing the efficiency of communication [13][31] Group 2: Design Process - Users can create complex designs by linking multiple images and providing comments for each, facilitating collaborative editing [22][25] - The process allows for precise adjustments without the need for extensive textual descriptions, streamlining the workflow [20][30] - The ability to visualize and modify designs in real-time significantly improves the creative process [29][33] Group 3: User Experience Transformation - The shift from user experience (UX) to agent experience (AX) positions Lovart as a collaborative partner rather than just a tool [29][30] - As users engage more with the agent, it learns their preferences, leading to a compounding effect in interaction efficiency [31][32] - Lovart's approach sets a new standard for creative design agents, emphasizing a seamless and interactive design experience [32][33]

Core Viewpoint - MiniMax Agent is positioned as a unique Vibe Coding product that simplifies the development process, allowing users to create complete web applications with a single command, covering front-end, back-end, and deployment functionalities [2][26]. Group 1: Product Features - Recent updates to MiniMax Agent include backend development deployment and scheduling capabilities, enhancing its functionality [2]. - The product allows users to create a fully functional e-commerce website with ease, demonstrating its versatility [3]. - Users can generate a comprehensive AI fortune-telling website, which includes features for long-term and short-term fortune calculations, as well as user management and payment capabilities [4][5]. Group 2: Development Process - The agent utilizes open-source projects for algorithm learning and employs simple random number generation for certain features [8]. - User information storage and payment integration are streamlined, with Supabase and Stripe being used for database management and payment processing, respectively [10][11]. - The agent conducts code testing and visual testing to ensure the functionality and integrity of the developed web applications [13]. Group 3: User Experience - The final product successfully integrates core functionalities, including fortune calculations, trial logic, and payment systems, with minor issues being promptly addressed [15]. - Users are provided with three trial opportunities before requiring login, enhancing user engagement [18]. Group 4: Market Implications - MiniMax Agent addresses common barriers faced by independent developers, such as backend development and payment integration, by abstracting these processes into simple commands [26]. - The product signifies a shift towards a future where cognitive understanding and problem-solving become the primary resources in business, rather than technical skills [27][28]. - The ultimate goal of technology is to empower individuals with ideas to participate in the commercial landscape [29].

Core Viewpoint - The article discusses the launch of Metaso's deep research feature, which is the first free product offering in the market that provides deep research capabilities, aiming to reduce costs while maintaining high accuracy in AI search and reasoning results [2][64]. Group 1: Deep Research Functionality - Metaso has implemented a segmented reinforcement learning approach to break down deep research tasks, significantly reducing resource consumption while ensuring high accuracy [3]. - The platform enhances user confidence by allowing them to verify information through various interactions and displays, effectively reducing model hallucinations [4][14]. - A novel interaction design reveals the dynamic "problem chain" during the execution of deep research tasks, providing users with insights into the model's reasoning process [7][14]. Group 2: User Interaction and Experience - The deep research results are presented in a more understandable format, utilizing various modalities to aid comprehension, such as audio explanations and interactive reports [15][21]. - Users can generate audio podcasts for each result, allowing for verification of information through listening [16]. - The platform highlights references and sources interactively, enhancing the user experience and understanding of the results [17]. Group 3: Case Studies and Applications - The article provides examples of how the deep research feature effectively addresses current social and financial issues, such as the inheritance dispute involving Wahaha's founder, Zong Qinghou, and the implications of stablecoins in the financial sector [27][40]. - A detailed timeline of events related to the inheritance dispute is presented, showcasing the platform's ability to organize and clarify complex information [30][33]. - The platform also offers strategic insights for gaming scenarios, demonstrating its versatility in handling various types of inquiries [50][61]. Group 4: Technological Innovation and Philosophy - Metaso's commitment to providing free access to advanced AI search and deep research services reflects a belief that the best technology should serve the most people [64][65]. - The company emphasizes that technological innovation is driven by the desire to reduce costs and improve service quality, rather than merely providing free services as a charitable act [64].

Core Viewpoint - The article discusses the capabilities of Kimi K2, a new model that has shown significant performance improvements in creating complex components for B-end applications, outperforming its predecessor, Claude Code [1][22]. Summary by Sections Kimi K2 Performance - Kimi K2 was tested immediately after its release, demonstrating strong capabilities even under increased difficulty by removing all code examples and design guidance, focusing solely on task requirements [2]. - The result was a comprehensive B-end component library featuring complex components such as calendar scheduling, step-by-step guide pop-ups, rich text editors, quick search components, filterable data tables, file tree components, and draggable data dashboard components [3]. Component Comparisons - A specific focus was placed on the draggable data dashboard component, which Kimi K2 handled effectively, while Sonnet 4 failed to deliver a functional version, highlighting K2's superior handling of edge cases and user interactions [4][5]. Component Details - The article outlines various components created using Kimi K2, including: - A customizable dashboard component allowing users to add, remove, and rearrange widgets [5]. - A file tree component displaying folders and file types with interactive features [7]. - A comprehensive calendar component for managing events and schedules [10]. - A modern rich text editor with a user-friendly formatting toolbar [11]. - An advanced data table component for structured data manipulation [13]. - A keyboard-driven quick operation center similar to tools used in popular applications [14]. API Integration and Usage - The article provides additional instructions for integrating Kimi K2 with Claude Code, addressing common issues users faced, such as API settings and environment variable configurations [16][17]. - It emphasizes the importance of using the correct API endpoints for domestic and international users [19][20]. Community Response and Impact - The release of Kimi K2 has generated significant discussion within the AI community, with researchers validating its capabilities and users sharing impressive use cases [22][24]. - The model's open-source nature has contributed to its rapid adoption and positive reception, contrasting with previous sentiments of stagnation in the AI industry [24].

Core Viewpoint - The K2 model, developed by Kimi, is a significant advancement in AI programming tools, featuring 1 trillion parameters and achieving state-of-the-art results in various tasks, particularly in code generation and reasoning [2][3][12]. Group 1: Model Capabilities - K2 has demonstrated superior performance in benchmark tests, especially in code, agent, and mathematical reasoning tasks, and is available as an open-source model [3][12]. - The model's front-end capabilities are comparable to top-tier models like Claude Sonnet 3.7 and 4, making it a strong contender in the market [4][16]. - K2's ability to integrate with Claude Code allows users to utilize its features without concerns about account bans, enhancing its practical usability [23][32]. Group 2: Cost Efficiency - K2 offers a competitive pricing structure, with costs as low as 16 yuan for one million tokens, making it significantly cheaper than other models with similar capabilities [34]. - The model's cost-effectiveness is expected to democratize access to AI programming tools in China, potentially leading to a surge in AI programming and agent product development [35][38]. Group 3: Future Implications - The introduction of K2 is anticipated to activate the potential of domestic AI programming products and agents, marking the beginning of a transformative phase in the industry [35]. - K2 fills a critical gap in the market by providing a practical and usable open-source model, which could lead to increased innovation and development in AI tools [34][36].