Keeping pace with news and developments in the real-time analytics and AI market can be a daunting task. Fortunately, we have you covered with a summary of the items our staff comes across each week. And if you prefer it in your inbox, sign up here!
NVIDIA launched NVIDIA Nemotron 3 Super, a 120‑billion‑parameter open model with 12 billion active parameters designed to run complex agentic AI systems at scale. Available now, the model combines advanced reasoning capabilities to efficiently complete tasks with high accuracy for autonomous agents.
The solution uses a hybrid mixture‑of‑experts (MoE) architecture that combines three major innovations to deliver up to 5x higher throughput and up to 2x higher accuracy than the previous Nemotron Super model. Features and capabilities include:
- A Hybrid Architecture: Mamba layers deliver 4x higher memory and compute efficiency, while transformer layers drive advanced reasoning.
- MoE: Only 12 billion of its 120 billion parameters are active at inference.
- Latent MoE: A new technique that improves accuracy by activating four expert specialists for the cost of one to generate the next token at inference.
- Multi-Token Prediction: Predicts multiple future words simultaneously, resulting in 3x faster inference.
Real-time analytics news in brief
Actian, the data and AI division of HCLSoftware, introduced the Actian AI Analyst (formerly Wobby), a conversational analytics solution featuring the Steward Agent that intelligently generates and provides ongoing maintenance of the semantic layer to speed time to value and deliver trusted, auditable insights. By combining AI-driven analysis with deterministic semantics, Actian AI Analyst allows organizations to scale analytics access, reduce business intelligence bottlenecks, and act on insights with confidence.
Amberd.ai announced the launch of its private, LLM-native decision system that lets enterprises ask plain-language questions and receive clear, sourced answers from their own data while protecting privacy and compliance. The solution allows multiple databases and documents to be queried in the context of industry-specific laws and regulations, providing a single, decision-ready answer organizations can trust. Amberd.ai’s platform is deployed in a private, enterprise-grade environment, ensuring that data remains under the customer’s control and is protected by strict governance and security controls.
Boomi announced new capabilities within the Boomi Enterprise Platform. Data activation brings data to life across systems and processes, delivering it with the right context and timing to power everything from AI to BI. The Boomi Enterprise Platform now adds new semantic context to help AI agents operate on grounded business realities, expands governed SAP data movement with change data capture, enhances transparency and oversight across agentic workflows, and introduces a dedicated European platform instance for localized data control.
CData Software announced major enhancements to CData Connect AI. The updates extend CData’s managed Model Context Protocol (MCP) platform with new capabilities across connectivity, context, and control, the three pillars required to move AI from experimentation to production. Additionally, this release gives teams the ability to build use-case-specific agent tools with the right business context, deploy them securely, and enforce granular controls over which data agents can access.
ChurnZero enhanced its embedded AI agents with connected knowledge source integrations, enabling them to more completely understand the nuances of any business, its products and services, the needs of its customers, and how its customer success team works. Available now, AI Knowledge Sources offers initial integrations with Atlassian’s Confluence and Zendesk Guide, with more integrations to follow shortly.
Clockwork.io announced the general availability of TorchPass Workload Fault Tolerance. Delivered as a core capability of the Clockwork.io FleetIQ platform, TorchPass applies the principles of Software-Driven AI Fabrics to distributed training, using Live GPU Migration to allow workloads to continue running through GPU failures, network disruptions, driver bugs, and even full node crashes, without checkpoint restarts or lost progress.
Conga unveiled AiMe, a unified AI layer providing prescriptive, agentic, and proactive intelligence to every customer workflow. Built directly into the Conga Advantage Platform, AiMe connects data, predicts next-best actions, automates tasks, and adapts in real time, helping organizations eliminate operational blind spots, accelerate decision-making, and improve commercial performance.
Corvex announced the early availability of Corvex Secure Model Weights, a patent-pending solution that enables Frontier AI model builders and enterprises to deploy inference workloads on third-party GPU infrastructure without exposing their model weights. The solution’s architecture leverages Hardware-based Trusted Execution Environments (TEEs). By utilizing NVIDIA’s Confidential Computing instructions, Corvex ensures that model weights remain cryptographically isolated and are only decrypted within the GPU’s secure silicon boundary, which renders them invisible even to the infrastructure provider.
CoreWeave announced CoreWeave Flexible Capacity Plans, including Flex Reservations and Spot, designed to match the dynamic reality of modern AI workloads. The solutions enable customers to move beyond the binary choice of reserved capacity versus on-demand capacity. Specifically, with the addition of Flex Reservations and Spot, CoreWeave introduces a unified capacity framework that allows customers to reserve what’s steady, protect what must be guaranteed, and shift interruptible work, aligning cost and certainty with real demand patterns.
Cribl announced its next generation of Cribl Search, an AI-native log intelligence solution built on an agentic telemetry architecture. Designed from the ground up for AI‑driven workloads, the new Search experience helps organizations cut log management costs, resolve security and IT incidents faster, and streamline tooling.
Crusoe announced Crusoe Edge Zones, powered by Crusoe Spark, a new solution that brings AI compute to virtually any location. Crusoe Edge Zones leverage Crusoe’s proprietary Crusoe Spark modular data centers to provide low-latency infrastructure and sovereign AI deployments for customers globally. With Crusoe Edge Zones, customers can deploy dedicated Crusoe Spark units in geographically targeted locations, unlocking use cases that traditional hyperscale and neo-cloud providers do not support.
Databricks launched Genie Code, an autonomous AI agent that changes how data work gets done. Genie Code can carry out complex tasks such as building pipelines, debugging failures, shipping dashboards, and maintaining production systems. Just as agentic coding tools have transformed software engineering, moving developers from autocomplete-style assistance to agent-driven development, Genie Code brings the same paradigm shift to data engineering, data science, and analytics.
Datadog announced that its MCP Server is generally available. For developers embedding AI agents into development and operational workflows, the Datadog MCP Server provides access to live observability data, so teams can debug in their preferred choice of AI coding agents or an Integrated Development Environment with real-time telemetry and take action within established security and governance controls.
Dataiku announced the launch of the Platform for AI Success, a strategic evolution of its enterprise AI platform designed to take enterprise AI from pilots into trusted and measurable business performance. The launch includes three products: Dataiku Agent Management for cross-platform agent governance and business impact validation; Dataiku Cobuild for AI-assisted agent building in a visual, inspectable environment; and Dataiku Reasoning Systems for industry-specific decision intelligence delivered by teams of agents.
Denodo announced the launch of Platform 9.4, which was built to address the challenge of AI systems not always having reliable access to trusted enterprise data in real-time. The new Denodo Platform 9.4 enables AI agents to securely access live enterprise data, both structured and unstructured, and reinforces Denodo’s role as the trusted data foundation for agentic and generative AI. The update also accelerates AI workloads and analytics through the Lakehouse Accelerator.
Equinix unveiled the Distributed AI Hub, powered by Equinix Fabric Intelligence, to provide a single, unified framework for enterprises to connect, secure, and simplify their distributed AI ecosystems. The Hub is a neutral location that allows enterprises to discover, connect to, and consume AI infrastructure providers, including model companies, GPU clouds, data platforms, network and security services, and AI frameworks, all through private, low-latency connectivity at Equinix’s 280 high-performance data centers.
FluidCloud announced the launch of its Large Infrastructure Model (LIM), an LIM purpose-built for true multi-cloud readiness. LIM generates production-grade infrastructure as code (IaC) with 99%+ accuracy and is purpose-built for multi-cloud architecture reasoning, cross-cloud Terraform migration, and operational continuity. Additionally, FluidCloud’s LIM provides core capabilities for enterprise-scale multi-cloud operations that drive faster, safer migrations with measurable savings and create a scalable foundation for long-term multi-cloud resilience.
Galileo announced the release of Agent Control, an open-source control plane that empowers organizations to define and enforce desired behavior across all their AI agents. Users can write policies once and deploy anywhere, essential for deploying AI in critical enterprise processes. Strands Agents, CrewAI, Glean, and Cisco AI Defense will be the first to integrate with Agent Control to help their customers quickly deploy centralized governance for agents.
Intel launched the Intel Core processor Series 2 with P-cores, an industrial-ready platform engineered for mission-critical edge applications. The processors address the critical challenges facing modern industrial operations, which demand processors that can handle multiple critical workloads simultaneously. Intel also announced its latest Edge AI suite for Health and Life Sciences, providing validated reference pipelines and benchmarking tools for AI-powered patient monitoring solutions.
KX announced the general availability of two capital markets agentic AI blueprints developed in collaboration with NVIDIA: an AI Research Assistant and Trading Signal Agents. Available now, the KX AI Research Assistant blueprint is designed to accelerate research workflows by improving retrieval, summarization, and synthesis across structured market data, unstructured content, and proprietary documents. The KX Trading Signal Agents blueprint is designed to help teams discover, validate, and monitor signals in real time by aligning multi-modal information in market time and producing governed outputs.
Mindbreeze announced significant enhancements to the Mindbreeze Insight Workplace, a central control plane for enterprises. These enhancements expand capabilities for Insight Touchpoints and Insight Journeys and enable organizations to scale agentic AI in a structured, governed, and highly repeatable way. New capabilities include pre-built, role-specific AI applications (Insight Touchpoints), automated multi-step workflows (Insight Journeys), and a centralized governance layer within the Insight Workplace.
OneTrust announced the expansion of its AI-Ready Governance Platform to include real-time monitoring and enforcement capabilities across agents, models, and data. The new capabilities allow organizations to extend AI governance into the deeper operational layers where models, agents, data, and infrastructure run, and give teams a practical control plane for governing agents in real time.
SurePath AI announced MCP Policy Controls, which provides real-time controls over what MCP servers and tools are allowed to be used. SurePath AI enforces an organization’s policies on exactly which MCP servers and tools are allowed by controlling local MCP hosts and their connections to local MCP servers. These policies can leverage built-in classifications of whether a tool is destructive or not, or be customized explicitly to each organization’s security requirements.
Teradata announced new agentic and multi-modal data capabilities for Teradata Enterprise Vector Store. Advanced new features include unstructured data integration; hybrid search that combines semantic and lexical search with metadata-driven techniques for more accurate, context-aware retrieval; multi-modal embeddings with support for text, image, and audio embeddings with richer semantic representations; higher embedding dimensions, and LangChain integration.
ThoughtSpot announced Spotter Semantics, an agentic semantic layer designed to deliver consistent, contextual, and trustworthy insights. Built on an AI-native foundation, Spotter Semantics transforms raw, fragmented data into governed business context that AI agents can reliably understand and act on. By serving as a context-aware translation engine between complex data sources and AI agents, Spotter Semantics ensures every natural language query results in an accurate, explainable, and actionable answer, at enterprise scale.
Tricentis announced the launch of its unified, agentic quality engineering platform powered by the new Tricentis AI Workspace. The AI Workspace operates as a single, unified command center with shared context, integrated workflows, and native agent-to-agent collaboration to serve as the system of record and ‘control tower’ for agentic quality engineering, coordinating AI agents across testing, automation, performance, and quality intelligence, while embedding governance, approvals, and auditability directly into execution.
Whatfix launched AI Roleplay training in Mirror, which makes Mirror an AI-first training platform designed to prepare customer-facing teams. With this launch, Mirror evolves beyond system simulation to combine adaptive AI-driven roleplay with realistic enterprise application simulations, enabling frontline teams to practice customer conversations and workflows together in a single, risk-free environment.
Partnerships, collaborations, and more
1NCE opened access for its customers to the LoRaWAN services of Netmore. Specifically, the Netmore LoRaWAN Network Server (LNS) Plugin provides 1NCE customers access to cellular and LoRaWAN IoT coverage options through one platform. The launch of the Netmore Plugin marks the beginning of strategic collaboration to expand the combined offering of the two biggest massive IoT network specialists, featuring 1NCE as the biggest private IoT network in the cloud and Netmore as the largest LoRaWAN network and platform provider in the world.
Amazon Web Services (AWS) and Cerebras Systems announced a collaboration that will, in the coming months, deliver the fastest AI inference solutions available for generative AI applications and LLM workloads. The solution, to be deployed on Amazon Bedrock in AWS data centers, combines AWS Trainium-powered servers, Cerebras CS-3 systems, and Elastic Fabric Adapter (EFA) networking. Later this year, AWS will also offer leading open-source LLMs and Amazon Nova using Cerebras hardware.
Cohesity announced a strategic integration with Datadogto deliver enterprise-grade AI Agent Resilience, combining continuous observability with rapid, automated data recovery for production AI environments. Through this integration, Datadog provides real-time monitoring across cloud infrastructure, object stores, AI workloads, and application services, enabling enterprises to establish behavioral baselines and detect anomalous activity.
Hammerspace announced a partnership with Secuvy to deliver a “Data-First” approach that turns raw data into secure AI outcomes. Together, the companies unify distributed unstructured data into a global namespace and continuously discover, classify, catalog, and control it across both on-premises and cloud. To that end, the two companies keep data continuously AI-ready as it changes, so governance and access controls stay current from proof-of-concept to production.
MariaDB plc announced that it has entered into a definitive agreement to acquire GridGain Systems. By merging MariaDB’s AI-ready relational database with GridGain’s scalable, in-memory power, MariaDB is setting a new industry standard: sub-millisecond data infrastructure for the agentic era. Specifically, the acquisition fuses MariaDB’sACID-compliant transactional integrity for sensitive data, native vector capability, and AI support with GridGain’s extreme-scale and in-memory processing that eliminates the disk-drive tax on performance.
Mirantis and Netris announced an integration that makes network automation and hard tenant isolation a native part of Kubernetes cluster delivery for neoclouds, telecom operators, and enterprises building AI infrastructure. The integration automates Kubernetes cluster delivery and data center networking for AI workloads, eliminating two of the biggest operational bottlenecks: the lack of a standardized path to cluster deployment and the manual, fragmented network provisioning processes that slow infrastructure rollout.
Precisely announced a new OEM partnership with Matillion, bringing cloud-native Extract Transform Load (ETL) capabilities to the Precisely Data Integrity Suite. The new offering enables organizations to move, transform, and prepare data within a single suite of interoperable services, eliminating reliance on fragmented tools and manual ETL processes. Additionally, by combining Matillion’s modern, cloud-native transformation platform with the broader Precisely Data Integrity Suite, organizations can modernize data environments faster and deliver AI-ready data to power AI, automation, and analytics initiatives.
If your company has real-time analytics news, send your announcements to ssalamone@rtinsights.com.
In case you missed it, here are our most recent weekly real-time analytics news roundups:
