#LargeLanguageModels

[nonetheless]

What lessons from the 16th century can tell us about AI and LLMs: "Methodical banality" @aeon.co (Plus- the Graphophone)

Honoring authenticity: https://roughlydaily.com/2025/05/04/when-i-use-a-word-humpty-dumpty-said-in-rather-a-scornful-tone-it-means-just-what-i-choose-it-to-mean-neither-more-nor-less/

[mysocial8onetech]

Learn how Qwen2.5, a large language model developed by Alibaba Cloud, revolutionizes AI with its ability to process long contexts up to 128K tokens and support over 29 languages. Pretrained on a large-scale dataset of 18 trillion tokens, it enhances high-quality code, mathematics, and multilingual data. Discover how it matches Llama-3-405B’s accuracy with only one-fifth of the parameters.

[cloudystructureshark]

Future Trend in Private Large Language Models

Future Trend in Private Large Language Models

As artificial intelligence rapidly evolves, private large language models (LLMs) are becoming the cornerstone of enterprise innovation. Unlike public models like GPT-4 or Claude, private LLMs are customized, secure, and fine-tuned to meet specific organizational goals—ushering in a new era of AI-powered business intelligence.

Why Private LLMs Are Gaining Traction

Enterprises today handle vast amounts of sensitive data. Public models, while powerful, may raise concerns around data privacy, compliance, and model control. This is where private large language models come into play.

A private LLM offers complete ownership, allowing organizations to train the model on proprietary data without risking leaks or compliance violations. Businesses in healthcare, finance, legal, and other highly regulated sectors are leading the shift, adopting tailored LLMs for internal knowledge management, chatbots, legal document analysis, and customer service.

If your enterprise is exploring this shift, here’s a detailed guide on building private LLMs customized for your business needs.

Emerging Trends in Private Large Language Models

1. Multi-Modal Integration

The next frontier is multi-modal LLMs—models that combine text, voice, images, and video understanding. Enterprises are increasingly deploying LLMs that interpret charts, understand documents with embedded visuals, or generate responses based on both written and visual data.

2. On-Premise LLM Deployment

With growing emphasis on data sovereignty, more organizations are moving toward on-premise deployments. Hosting private large language models in a secure, local environment ensures maximum control over infrastructure and data pipelines.

3. Domain-Specific Fine-Tuning

Rather than general-purpose capabilities, companies are now investing in domain-specific fine-tuning. For example, a legal firm might fine-tune its LLM for case law analysis, while a fintech company might tailor its model for fraud detection or compliance audits.

4. LLM + RAG Architectures

Retrieval-Augmented Generation (RAG) is becoming essential. Enterprises are combining LLMs with private databases to deliver up-to-date, verifiable, and domain-specific responses—greatly improving accuracy and reducing hallucinations.

Choosing the Right LLM Development Partner

Implementing a secure and scalable private LLM solution requires deep expertise in AI, data security, and domain-specific knowledge. Collaborating with a trusted LLM development company like Solulab ensures that your organization gets a tailored solution with seamless model deployment, integration, and ongoing support.

Solulab specializes in building enterprise-grade private LLMs that align with your goals—whether it’s boosting customer experience, automating workflows, or mining insights from unstructured data.

Final Thoughts

The future of enterprise AI lies in private large language models that are secure, customizable, and hyper-efficient. As businesses look to gain a competitive edge, investing in these models will no longer be optional—it will be essential.

With advancements in fine-tuning, multi-modal intelligence, and integration with real-time data sources, the next generation of LLMs will empower enterprises like never before.

To stay ahead in this AI-driven future, consider developing your own private LLM solution with a reliable LLM development company like Solulab today.

[elenajhonson]

The power of Custom Generative AI: Transforming how businesses communicate

The role of AI in modern business is no longer just experimental—it’s essential. One of the most powerful tools driving this transformation is Generative AI, especially when paired with custom-built large language models (LLMs).

At Imobisoft, they specialise in developing bespoke LLM and Generative AI solutions that are tailored to each client's specific needs. These solutions are changing the way organisations operate—making communication smarter, content faster to produce, and customer interactions more meaningful.

Why Custom LLMs Outperform One-Size-Fits-All Tools

Unlike general-purpose AI platforms, custom LLMs can be trained using your company’s own data, vocabulary, tone, and intent. This means your AI doesn’t just respond—it understands. Whether you're dealing with internal communications, customer support, or content creation, these tools adapt to your business like a true team member would.

AI-Powered Content That Scales

Writing blogs, emails, reports, or social posts can eat up time and resources. Custom Generative AI models streamline this process by creating high-quality, on-brand content instantly. Teams can now focus more on strategy and creativity, leaving routine writing tasks to intelligent automation.

Engage Customers Like Never Before

Customer service is evolving. With AI-driven solutions built on LLMs, businesses can deliver consistent and contextual interactions at scale—day or night. From chatbots to smart email responders, AI ensures no customer is left waiting.

Future-Ready and Fully Scalable

The landscape of technology is changing rapidly, and businesses that integrate advanced AI tools today are the ones that will lead tomorrow. Their solutions are designed with scalability in mind—so they grow with your organization.

Custom LLM and Generative AI applications are more than just tech trends—they're tools for real, measurable progress. At Imobisoft, they make these tools accessible, powerful, and aligned with your goals.

Let’s build the future of your business together—with intelligence that thinks like you do.

[govindhtech]

Structured Object Language Model: Lightweight Language Model

Structured OLM

Amazon's lightweight NLP approach, Structured Object Language approach (SoLM), generates structured objects within schemas. Discover how self-supervised denoising and CABS reduce hallucinations.

One of the most important features of today's generative models is their ability to convert unstructured, partially unstructured, or poorly structured inputs into structured objects that follow schemas like relational-database fixed schemas, document store flexible schemas, function signatures, API specifications, etc.

Large language models (LLMs) can do this work if given schema requirements and processing instructions. Most LLMs today have a JSON mode or structured-outputs mode to safeguard users from prompt engineering.

This method has various limitations. First, LLMs are expensive when expanding to databases with millions or billions of entries or requests. Second, prompt engineering can be complicated. Third, the built-in structured-outputs and JSON modes can only support so many schemas.

In Empirical Methods in Natural Language Processing (EMNLP), Amazon published a lightweight Structured object language Model (SoLM) on ArXiv to solve this problem locally. Unlike general-purpose LLMs, structured object language models are trained to produce objects only within a schema. SoLM's achievements include self-supervised denoising, a unique training method, and confidence-aware substructure beam search (CABS), a decoding method for inference time that lowers hallucinations.

In tests, Structured object language Model had an order of magnitude higher cost efficiency than state-of-the-art LLMs and comparable output accuracy. Found that CABS decoding outperformed beam search decoding in product attribute generation recall by 16.7% at 90% accuracy.

Applications

The structured-output paradigm unites seemingly unconnected AI/ML difficulties in study. A issue could arise if the structured entity has multiple aspects or redundant, interconnected information. A brief, type-constrained structured data may be part of the object, along with a long, natural language description text.

Multidimensional objects with descriptive and key attribute listings are often used to list things, homes, occupations, etc. Structured object language Model lets you design an object with absolute world knowledge consistency and relative object consistency.

Typically, a structured-output model is fed unstructured data and allowed to produce a structured object. Amazon suggests a self-regenerating machine using Structured object language Model in study. It just gives the model a schema-structured object and lets it replicate it.

Instead of structuring, clean, normalise, correct, and/or finish the input while making it self-consistent. Input can contain extra unstructured content, a structured record, or a record with a different schema. Structured object language Model always produces a clean record according to the schema, regardless of input.

The self-regenerating machine can repair incorrect facts, normalise unnormalized facts, complete missing descriptions, and correct inaccurate descriptions. Since these occupations are interconnected, doing them separately produces dependence cycles (e.g., should one develop descriptions based on facts or extract facts from descriptions?). These dependencies are best solved organically by self-regeneration.

Innovations

Amazon trains Structured object language Model with self-supervised denoising. The idea is to use any sample of things from a database, add false noise, then train the model to restore their original shapes. What Amazon feeds the model improves in quality. More aggressive noise, such as destroying the object's structure or randomly rearranging tokens, teaches the model to work with unstructured input and improves object quality.

Even if they are programmed to offer the most likely next token in a series, LLMs use multiple decoding algorithms to determine outputs at inference time. One of the most common is beam search decoding, where the model parallelises multiple candidate sequences and chooses the best cumulative probability. Greedy decoding simply chooses the token with the highest probability at each round, therefore it cannot guarantee the highest-probability sequence across a set number of turns. The beam width represents the number of sequences the model considers.

Structured object language produce Models are collections of key-value pairs, where the value is the type's value, such as an item's brand, and the key is a schema data type, such as “brand” in product listings. Amazon uses tokens (“” and “”) to identify keys and values.

The key-value pair, not the token, is the atomic component of confidence-aware substructure beam search. The key-value pair's likelihood can be calculated from the LLM's output confidence level. Another experiment used an independently trained confidence score model that used one of the LLM's inner layers' intermediate representation. This worked better than using model confidence ratings alone.

Amazon proves a 7 billion parameter Structured object language Model model equals or exceeds prompt-engineering methods on considerably bigger foundational models in fact completeness, accuracy, and descriptive content quality and factuality. CABS decoding greatly improves fact accuracy by removing hallucinated facts.

[craigbrownphd]

OpenSearch 3.0 augments vector database performance, search infrastructure, and more

#GenAI #DataBricks #DataScience https://www.kmworld.com/Articles/ReadArticle.aspx?ArticleID=169387&utm_source=dlvr.it&utm_medium=tumblr

[spookywhisperomen]

AI-Powered Future: From Machine Learning to Avatars & Co-Pilots

Artificial Intelligence (AI) is no longer a visionary term—it's already revolutionising sectors of the world today. From AI building and machine learning building to AI as a service, companies are leveraging bleeding-edge technologies to remain ahead of competition and innovate at a quicker rate. With the changing environment, recruiting talented experts like AI engineers and ChatGPT developers has become crucial. Let's get into how these innovations, particularly in industries such as retail, are dictating the future with enterprise AI solutions, large language model creation, AI co-pilot creation, and AI avatar creation.

The Expanding Scope of AI Development  

An AI development company deals with the creation of intelligent systems that are adept at tasks that have traditionally been performed by people. It is the field that has a rich collection of information, like problem-solving, decision-making, natural language understanding, and learning from data, as its central issues.

AI development today encompasses not just machine learning but also natural language processing, computer vision, and robotics, resulting in a proliferation of powerful AI apps enabling organizations to automate processes, improve customer service, and uncover business insights.

Machine Learning Development: A Pillar of AI Innovation  

A machine learning development represents the central operational element for present-day AI environments. The organization focuses on creating intelligent data-based systems that achieve performance improvement through learning instead of requiring manual development for each new function.

The company use extensive datasets to develop models that adjust to actual operating conditions and produce precise and efficient  and scalable AI solutions for complicated enterprise issues. Modern AI solutions depend on machine learning development to create predictive analytics and recommendation engines and real-time decision-making systems that power contemporary enterprise operations.

When you work with an established machine learning development company, your business receives the necessary resources to establish strong AI capabilities. These solutions provide the tools needed for competitive advantage and fast innovation and operational readiness across healthcare, finance, and machine learning in retail environments.

AI as a Service: Democratizing AI Access  

The AI delivery sector experiences a profound transformation through the establishment of Artificial Intelligence as a Service (AIaaS).  Organizations at any scale can access advanced AI technology through cloud platforms, which eliminates the requirement for large initial expenses in infrastructure or personnel. Organizations that subscribe to AI services gain the capability to add natural language processing together with image recognition and predictive analytics and conversational AI to their system or operation without difficulty. This transformation enables companies without the means to create internal AI development teams to access AI technology, thus extending the advantages of artificial intelligence to multiple sectors.

Why Hire AI Engineers and ChatGPT Developers?  

As AI becomes more pervasive, the demand for specialized talent is soaring. Hiring artificial intelligence engineers skilled in machine learning, data science, and algorithm design is crucial for companies aiming to build custom AI solutions that align with their unique business goals.

Similarly, hiring ChatGPT developers—experts in large language model development—is essential for companies seeking to integrate advanced conversational AI into their customer service, marketing, or internal workflows. These developers tailor AI chatbots and virtual assistants that understand and respond naturally to human language, enhancing user engagement and operational efficiency.

Machine Learning in Retail: Revolutionizing the Shopping Experience  

Machine learning in retail technologies drives substantial changes in the retail sector together with other industries. The retail sector deploys machine learning, which generates individualised customer interactions alongside predictive sales patterns and efficient stock handling and fraud prevention.

Through extensive customer data analysis, machine learning algorithms detect purchasing behaviour and individual preferences, which retailers leverage to create precise promotions and personalized product suggestions. This simultaneous effect increases both revenue and customer dedication.

The retail industry implements machine learning to improve supply chain management operations, which enables efficient product availability while decreasing both waste and expenses. AI-driven market insights empower retailers to fast-track their responses to consumer needs and market trends, which protects their competitive position.

Enterprise AI Solutions: Scaling Intelligence Across Organizations  

Large corporations are more and more using enterprise AI solutions to simplify tough processes, boost their decision-making, and discover new sources of income. These are usually a mix of AI technologies, that may include such versions as machine learning, natural language processing, and robotic process automation, inside a single platform that cares for every business function.

A definite example in favour of this is that from predictive maintenance in manufacturing to detecting fraud in banking, enterprise AI solutions become those drivers which support this efficiency and, in some cases, the process of innovation. To leverage their AI to reach full potential, firms often invest in the development of huge language models to get their AI to understand human-like text and make better communication and insights possible.

The Rise of AI Co-Pilots and AI Avatars  

The AI Co-Pilot Development and AI Avatar Development are currently the trendiest sectors of the AI industry.

AI Co-Pilot Development: AI co-pilots function as smart helpers, who aid experts in handling their assignments in complex conditions. Be it writing software codes, guiding pilots in their navigation, or assisting customer service agents, AI co-pilots do all this and even more. These AI-powered friends never stop learning; they change according to the user's preferences and give their human colleagues contextual insights, so in this way, they revolutionise work in every existing industry.

AI Avatar Development: AI avatars are the new age of amazing virtual assistants, backed by high-level AI. They employ the power of natural language processing, computer vision, and emotion recognition to establish a conversational connection with users and also make themselves a part of the user's life. Whether it is virtual customer care reps or personalized health coaches or hosts for entertainment, AI avatars inject human-like touch in the world of automation, thus creating more engaging experiences for people.

Large Language Model Development for Scalable AI Solutions  

Large language model development is like the infrastructure on which modern AI runs. In sum, it is large language model development that allows machines to understand and generate human-like text in bulk, thereby making communication more human-like. This trend touches every major and minor AI-driven innovation and contributes to such principles as personalization, productivity, and innovation.

Final Thoughts  

For businesses that want to do well with this AI-powered future, the investment in artificial intelligence development and artificial intelligence as a service is not something that is optional any more; it's essential. Employing artificial intelligence engineers and ChatGPT developers guarantees that you have the right skills to develop and deliver AI solutions that are at the cutting edge of technological innovation.

Osiz Technologies creates intelligent AI solutions that help businesses innovate and grow across various industries. Our expert team builds advanced tools like virtual assistants and automation systems to prepare your business for the future.

[ailatestupdate]

DeepSeek Prover-V2 671B stands at the forefront of AI-driven formal mathematics.

[tsqc]

AI Leadership & Ethical Innovation: Shaping the Future of Business

[totoshappylife]

Speculative Decoding for Verilog:

Excerpt from PDF: Speculative Decoding for Verilog: Speed and Quality, All in One Changran Xu1,3,†, Yi Liu1,3,†, Yunhao Zhou1,3, Shan Huang2,3, Ningyi Xu2, and Qiang Xu1,3 1The Chinese University of Hong Kong, Shatin, Hong Kong S.A.R. 2Shanghai Jiao Tong University, Shanghai, China 3National Technology Innovation Center for EDA, Nanjing, Jiangsu, China Abstract—The rapid advancement of large…

[meiiaiinc]

Mei AI is a global leader in AI solutions, offering industry-trained Large Language Models that can be tuned accordingly with company-specific data and hosted privately or in your cloud.

Our RAG ( Retrieval Augmented Generation ) based AI approach uses Embedded Model and Retrieval context ( Semantic Search ) while processing a conversational query to curate Insightful response that is specific for an Enterprise. Blended with our unique skills and decade long experience we had gained in Data Analytics solutions, we combine LLMs and ML Algorithms that offer great solutions for Mid level Enterprises.

We are engineering a future that allows people, businesses, and governments to seamlessly leverage technology. With a vision to make AI accessible for everyone on the planet, our team is constantly breaking the barriers between machines and humans.

[emergysllc]

How can verticalization of LLM pivot industries to increase accuracy, efficiency, and relevance in practical applications? Get insights here: https://lnkd.in/dW7B4iq5

Learn how tailored LLMs can empower professionals to leverage AI in meaningful, context-aware ways across sectors.

[yourtechdietblog]

Understanding the Power of Large Language Models (LLMs)

Large Language Models (LLMs) are advanced AI systems that can understand and generate text with remarkable accuracy. Trained on massive datasets, these models use transformer-based architectures to perform tasks like answering questions, summarizing content, and facilitating conversational AI.

✨ What Makes LLMs Special?

Everyday Applications: Tools like ChatGPT and Google Bard transform industries like education, healthcare, and customer service.

Advantages: LLMs enhance automation, boost productivity, and enable seamless human-AI interactions.

Limitations: Challenges include bias, high computational costs, and lack of true reasoning.

LLMs are shaping the future of AI by bridging the gap between human creativity and machine efficiency.

[ai-network]

Writer Unveils Self-Evolving Language Models

Writer, a $2 billion enterprise AI startup, has announced the development of self-evolving large language models (LLMs), potentially addressing one of the most significant limitations in current AI technology: the inability to update knowledge post-deployment.

Breaking the Static Model Barrier

Traditional LLMs operate like time capsules, with knowledge frozen at their training cutoff date. Writer's innovation introduces a "memory pool" within each layer of the transformer architecture, enabling the model to store and learn from new interactions after deployment.

Technical Implementation

The system works by incorporating memory pools throughout the model's layers, allowing it to update its parameters based on new information. This architectural change increases initial training costs by 10-20% but eliminates the need for expensive retraining or fine-tuning once deployed. This development is particularly significant given the projected costs of AI training. Industry analyses suggest that by 2027, the largest training runs could exceed $1 billion, making traditional retraining approaches increasingly unsustainable for most organizations.

Performance and Learning Capabilities

Early testing has shown intriguing results. In one mathematics benchmark, the model's accuracy improved dramatically through repeated testing - from 25% to nearly 75% accuracy. However, this raises questions about whether the improvement reflects genuine learning or simple memorization of test cases.

Current Limitations and Challenges

Writer reports a significant challenge: as the model learns new information, it becomes less reliable at maintaining original safety parameters. This "safety drift" presents particular concerns for customer-facing applications. To address this, Writer has implemented limitations on learning capacity. For enterprise applications, the company suggests a memory pool of 100-200 billion words provides sufficient learning capacity for 5-6 years of operation. This controlled approach helps maintain model stability while allowing for necessary updates with private enterprise data.

Industry Context and Future Implications

This development emerges as major tech companies like Microsoft explore similar memory-related innovations. Microsoft's upcoming MA1 model, with 500 billion parameters, and their work following the Inflection acquisition, suggests growing industry focus on dynamic, updateable AI systems.

Practical Applications

Writer is currently beta testing the technology with two enterprise customers. The focus remains on controlled enterprise environments where the model can learn from specific, verified information rather than unrestricted web data. The technology represents a potential solution to the challenge of keeping AI systems current without incurring the massive costs of regular retraining. However, the balance between continuous learning and maintaining safety parameters remains a critical consideration for widespread deployment. Read the full article

[elenajhonson]

Unlocking business potential with LLM & Generative AI d

In today’s fast-paced digital environment, businesses are increasingly turning to artificial intelligence solutions to streamline operations, improve customer experiences, and gain a competitive edge. Among the most impactful technologies are Large Language Models (LLMs) and Generative AI tools, which are being adopted across industries to drive business automation, enhance creativity, and support smarter decision-making.

A leading technology service provider is playing a key role in delivering tailored AI development services by combining both proprietary and open-source models. Their expertise lies in building and integrating LLMs that align with specific business goals—whether it’s improving customer support, enhancing content creation, or simplifying data analysis through Natural Language Processing (NLP) solutions. These AI systems are designed not only to automate repetitive tasks but also to provide valuable insights and personalised experiences.

The development process begins with a detailed understanding of the client’s objectives and challenges. Through workshops and data analysis, they identify the areas where AI can add the most value. Based on this discovery phase, a custom AI strategy and roadmap is created, complete with ethical guidelines, system design, and clear success metrics.

Once the strategy is in place, a prototype is developed and tested in real-world conditions. Feedback from users and stakeholders is used to refine the system for better accuracy, usability, and fairness. Following successful testing, the solution is scaled up and fully integrated into existing business workflows. Training, documentation, and continuous support are provided to ensure smooth adoption and long-term performance.

Their services cover a wide range of AI-driven capabilities, including intelligent virtual assistants, AI-powered data analytics, predictive modelling, content generation, and seamless IoT and AI integration. These solutions are not only built for current needs but are also designed to evolve alongside the business, with regular model updates and performance monitoring.

What sets this approach apart is the emphasis on ethical AI development, user testing, and scalable architecture. By focusing on measurable results and long-term impact, this provider helps organisations confidently embrace digital transformation with AI and unlock new growth opportunities through advanced AI technologies.

Businesses exploring AI for workflow automation, customer engagement, or data analysis can benefit greatly from such a structured, human-centered approach to LLM & Generative AI Development. It’s a forward-thinking investment that prepares companies to thrive in an AI-powered future.

[govindhtech]

Announcing LangChain Postgres open-source Improvements

Open-source LangChain PostgreSQL upgrades

Google Cloud contributed heavily to the library and updated LangChain Postgres at Google Cloud Next ’25. These upgrades enable all application developers to design database-backed agentic gen AI solutions utilising open source technologies.

LangChain, an open-source framework, simplifies agentic gen AI systems that use massive language models. It connects large language models (LLMs) to other data sources for more powerful and context-aware AI applications. LangChain regularly interacts with databases to efficiently manage and extract structured data. The langchain-postgres package integrates PostgreSQL databases to load documents, store chat history, and store vectors for embeddings. Connectivity is needed for LLM-powered apps to use relational data, perform semantic searches, and generate memory chatbots.

Google Cloud enhancements include enterprise-level connection pooling, faster SQL filtering with relational metadata columns, and optimised performance with asynchronous PostgreSQL drivers. It also included:

Developers can use LangChain to create vector databases with vector indexes.

Flexible database schemas for more robust and manageable applications

For better security, the LangChain vector store APIs follow the least privilege principle and clearly distinguish database setup and usage.

Some new enhancements

Improved security and connectivity

Developing secure and dependable generative AI systems requires careful consideration of how your application interacts with the data architecture. Its LangChain Postgres contributions have prioritised security and connection through several key changes.

Following the least privilege concept has been our focus. The revised API distinguishes between database schema creation and application use rights. This separation lets you restrict the application layer's database schema changes. Separating these tasks can boost AI application security and reduce the attack surface.

Maintaining a pool of database connections reduces the overhead of making new connections for each query. This stabilises your application by efficiently limiting resource utilisation and preventing thousands of idle PostgreSQL connections. It also improves speed, especially in high-throughput scenarios.

Designing schema better

The langchain-postgres package historically only allowed schemas with fixed table names and a single json metadata column to resemble vector databases. PostgreSQL's sophisticated querying features allow you to filter non-vector columns to improve vector search quality. Our LangChain postgres package modifications let you define metadata columns to combine vector search queries with SQL filters when querying your vector storage.

Use the new LangChain PostgreSQL package to turn your PostgreSQL database structure into an AI workload with a few lines of code. This eliminates data schema migration.

Features ready for production

Google Cloud introduced vector index management and first-class asynchronous driver integrations in LangChain to enable production-scale applications. Asynchronous drivers enable non-blocking I/O operations, improving performance. This helps your application grow efficiently, reduce resource consumption, and increase responsiveness to handle more concurrent requests.

LangChain may now directly create and maintain vector indexes. This lets you utilise LangChain to describe and build your entire application stack, from database schema to vector index creation, using an infrastructure-as-code technique for vector search. This end-to-end connection simplifies development and makes LangChain AI-powered apps easy to set up and manage by using asynchronous operations and vector search.

LangChain packages for Google Cloud databases were upgraded by Google Cloud. It upstreamed those changes from its packages into LangChain PostgreSQL so developers on any platform could use them. Generative AI applications increasingly rely on databases, therefore software libraries must offer high-quality database connectors to exploit your data. These databases root LLMs, provide RAG application knowledge, and fuel high-quality vector search.

Get started

A quickstart application and langchain-postgres package are available now! Use this guide to switch from the old langchain-postgres package to Google's. Use AlloyDB's LangChain package and Cloud SQL for PostgreSQL to use GCP-specific capabilities like AlloyDB AI's ScaNN index. Create agentic apps with MCP Toolbox.