language models Archives

Tech Optimizer

March 17, 2026

EDB Accelerates Postgres for the Agentic Workforce Era with NVIDIA AI Infrastructure and Interactive Analytics

EnterpriseDB (EDB) has advanced its integration with NVIDIA's cuDF for Apache Spark to enhance Postgres® performance on NVIDIA AI infrastructure, achieving analytics capabilities up to 100 times faster than traditional methods. EDB emphasizes the need for a real-time analytics framework to address challenges posed by fragmented data silos and inefficient analytics processes. Key features of EDB Postgres AI include GPU-acceleration for interactive analytics, NVIDIA NIM model serving, fully air-gapped support for private registries, and high-speed retrieval with NVIDIA NeMo Retriever. Research indicates only 13% of enterprises have transitioned to production-scale agentic deployments, which report five times higher ROI. The EDB PG AI Analytics Engine can achieve 50–100x faster analytics on large datasets, supports lakehouse architectures, and ensures workload isolation and governance. EDB PG AI is positioned as a secure, compliant, and scalable platform for operationalizing data and AI workloads.

TrendTechie

March 12, 2026

Meta утверждает, что раздача пиратских книг через торренты подпадает под определение “добросовестного использования”

Meta is facing a class-action lawsuit from authors over the use of pirated books for training its Llama language model, arguing that the distribution of these books via BitTorrent constitutes fair use. The company used shadow libraries like Anna's Archive to gather text, and a California court has partially ruled that using pirated books for training large language models falls under fair use, although the issue of copyright infringement related to downloading and distribution via BitTorrent remains unresolved. The plaintiffs claim Meta has not previously indicated a fair use defense regarding file-sharing, while Meta asserts that authors have not shown evidence of their works being reproduced by the Llama model. The judge will ultimately decide on the admissibility of Meta's defense.

AppWizard

March 6, 2026

Google will now show which AI models are best at building Android apps

Google has introduced Android Bench, a tool for assessing AI model performance in Android app development. The top performer is Gemini 3.1 Pro, scoring 72.2%, followed by Claude Opus 4.6 at 66.6% and GPT 5.2 Codex at 62.5%. The benchmark evaluates models through real-world Android coding challenges with task completion rates between 16% and 72%. Google aims to facilitate the creation of Android applications from user prompts and has made the benchmark's methodology and tools available on GitHub.

AppWizard

March 6, 2026

Google says these AI models are best at coding Android apps

Google is testing various AI models for Android app development through a new platform called “Android Bench,” which evaluates the performance of leading AI language models (LLMs) against benchmarks specific to Android development. The benchmarks assess capabilities in areas such as Jetpack Compose, asynchronous programming, data persistence, dependency injection, navigation migrations, Gradle/build configurations, and interaction with Android components. Google has identified Gemini 3.1 Pro Preview as the top-performing model with a score of 72.4%, followed by Claude Opus 4.6 at 66.6% and OpenAI’s GPT 5.2 Codex at 62.5%. Gemini 2.5 Flash scored the lowest at 16.1%.

AppWizard

March 4, 2026

Google stuffs Gemini into Android Studio Panda 2 to build apps from prompts

Google has released Android Studio Panda 2, featuring an AI agent that generates applications and an AI-enhanced version upgrade assistant. This version is based on JetBrains IntelliJ IDEA's community edition. The AI capabilities are powered by Gemini, Google's large language models, with a free tier offering a lightweight version of Gemini 2.5 Pro. Developers can create prototypes with a single prompt, and the AI agent automates project planning, code generation, error analysis, and self-correction. Users must sign into Gemini and enable AI integration, with data collection practices in place. A demonstration showed the AI generating a bridge deal analyzer, which functioned but had inaccuracies in the generated code. Android Studio also experienced performance issues and deprecated certain features, including the Custom View preview and 3D mode in the layout inspector.

Tech Optimizer

February 14, 2026

PostgreSQL on Azure supercharged for AI

Over 80% of developers now incorporate AI tools into their workflows. PostgreSQL is preferred by 78.6% of developers engaged in AI and real-time applications. Microsoft has enhanced its PostgreSQL managed services to meet contemporary developer needs, contributing over 500 commits to the open-source project. The Azure Database for PostgreSQL supports both lift-and-shift and new open-source workloads, while the newly introduced Azure HorizonDB is designed for AI-native workloads. Developers can provision PostgreSQL instances directly from Visual Studio Code, and GitHub Copilot assists in writing and optimizing SQL queries. Azure Database for PostgreSQL facilitates integration with Microsoft Foundry for AI applications and supports high-performance similarity search with DiskANN vector indexing. PostgreSQL 18 is now available on Azure, offering enhanced performance and scalability. Nasdaq has modernized its Boardvantage platform using Azure Database for PostgreSQL and Microsoft Foundry to integrate AI for governance tasks. Azure HorizonDB is a fully managed PostgreSQL-compatible service designed for AI-native workloads, currently in private preview.

AppWizard

January 29, 2026

‘Secret’ car button can save lost drivers without opening Google Maps…

News Group Newspapers Limited has reminded users of its strict policies against automated access to its content, prohibiting data mining and activities related to artificial intelligence, machine learning, or large language models. For commercial use inquiries, users can contact crawlpermission@news.co.uk. Legitimate users mistakenly flagged by the system are encouraged to reach out to customer support at help@thesun.co.uk for clarification.

Tech Optimizer

January 27, 2026

EDB Announces Industry-First Book for Building Sovereign Data and AI Platforms with PostgreSQL

EnterpriseDB (EDB) has released a publication titled "Building a Data and AI Platform with PostgreSQL," authored by experts in PostgreSQL and data platforms. The book aims to guide executives and architects in transitioning generative and agentic AI into production-ready platforms. Research from EDB shows that while 95% of organizations plan to establish AI platforms in the next three years, only 35% currently use PostgreSQL for complex workloads, with just 13% achieving success at scale. The book emphasizes the importance of foundational platform design for successful AI deployment and offers a framework for establishing a platform mindset, identifying necessary architecture and governance, and mitigating data challenges. It has received endorsements from industry leaders, highlighting its practical guidance for moving AI from pilot projects to production. The book is available for purchase and will be distributed at the NVIDIA GTC 2026 conference.

Winsage

January 1, 2026

Windows Reborn: Microsoft Moves Copilot into the Kernel, Launching the Era of the AI-Native OS

Microsoft is transitioning its Windows operating system to an "AI-native" platform, embedding AI capabilities directly into the Windows kernel, marking a significant architectural shift not seen in three decades. This new approach, called the "Agentic OS," allows AI to manage files, system settings, and workflows proactively. The updated kernel, partially rewritten in Rust, includes a new NPU-aware scheduler that treats the Neural Processing Unit as a primary resource. Microsoft has introduced "Agent Workspace" and "Agent Accounts" for autonomous agents, ensuring actions are logged and audited for compliance. Communication between agents and the system is facilitated by the Model Context Protocol (MCP). Hardware requirements for the new OS have increased, with benchmarks set for NPUs achieving 80 to 100 TOPS. Major PC manufacturers are adjusting their portfolios to accommodate "Agentic PCs." The competitive landscape is evolving, with companies like Alphabet and Apple developing their own AI-native platforms. The introduction of the AI-native kernel raises privacy and security concerns, with Microsoft implementing measures to restrict third-party access to the kernel. Future updates may include "self-healing" capabilities and "Cross-Device Agency," leading to a more integrated personal AI experience.

AppWizard

December 30, 2025

Gemini control to Android: Will we talk to our phones more in 2026?

At the I/O 2025 event, Google presented Project Astra, showcasing its AI, Gemini, which can control Android devices to perform tasks like retrieving web content, playing YouTube videos, managing emails, and making phone calls. The demonstration illustrated Gemini's ability to navigate PDFs and transition between apps. Google introduced a Computer Use model for developers, enabling Gemini to interact with user interfaces in a human-like manner, currently optimized for web browsers. Apple is also enhancing Siri's capabilities to perform actions across multiple apps using voice commands. Google's approach is more generalized and not reliant on prior integrations, aiming to improve user interaction. The new Google Assistant, introduced in 2019, promised to allow voice operation of phones and multitasking across apps but faced challenges and limited adoption. Advancements in generative AI may enable more conversational command issuance, potentially addressing previous limitations. The integration of this technology could significantly impact wearable devices, allowing for phone control and information relay from secondary devices. The future of voice control as a primary interaction method for smartphones remains uncertain.