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!
Snowflake announced dynamic model routing within Cortex AI Gateway and Snowflake’s flagship AI products, alongside expanded access to leading open models. With the addition of dynamic model routing, Cortex AI Gateway can automatically select the specific model with the optimal balance of quality and cost for the task at hand. Dynamic model routing is also integrated across Snowflake’s flagship AI products, including Snowflake CoCo and Snowflake CoWork, and is available to third-party AI agents using Cortex AI Gateway.
This new capability directs lower-complexity or repetitive tasks to more efficient models, while work that requires deeper reasoning is routed to frontier models. This helps customers reduce unnecessary inference spend, without having to manage model selection for each request themselves. Snowflake will also be expanding customers’ access to leading open models, including DeepSeek-V4-Flash 0731 and GLM-5.3, through Snowflake Cortex AI. This adds to Snowflake’s extensive model library, giving customers even more options to balance model quality and cost, while keeping governed data secure within Snowflake.
Real-time analytics news in brief
MLCommons announced the release of MLPerf Client v2.0, the latest version of its industry-standard benchmark for evaluating AI performance on personal computers. Version 2.0 introduces a range of new features, refinements, and expanded support designed to keep pace with the rapid evolution of AI-enabled hardware and software. Notable changes include:
- New Image Generation Category: Adding this category significantly enhances the benchmark’s utility by enabling evaluation of generative visual capabilities. This category features Flux.2 klein 4B as an experimental test.
- New Agentic AI Category: Benchmarks agentic AI performance through Software Engineering (SWE) Agent and Data Analyst Agent scenarios. It reports end-to-end performance, including a breakdown of LLM inference and tool execution times.
- Updated LLM Inference tests: This release upgrades Phi 3.5 mini instruct in the required workloads to Phi 4 Mini Instruct, while introducing Qwen 3 8B as an experimental test. Base tasks now include an Intermediate Summarization task featuring an input prompt of roughly 4K tokens.
Adronite announced the launch of Codistry, an AI-powered coding platform. At the core of Codistry is the Adronite Context Engine, or ACE, Adronite’s patent-pending technology that builds a live relational understanding of a software codebase and provides AI models with precisely the right context needed to complete each task. The solution is designed to help engineering teams develop, understand, and maintain complex software with greater accuracy, lower costs, and more control over their development environments.
Alteryx announced the general availability of Alteryx One: Google Edition on Google Cloud Marketplace. Google Edition brings Alteryx’s in-place analytics experience to Google Cloud BigQuery, enabling organizations to analyze data where it resides while giving business users a simpler, governed way to work directly with BigQuery data and take advantage of Google Cloud’s built-in AI capabilities through Alteryx’s familiar no-code experience. It also simplifies procurement, deployment, and management through Google Cloud Marketplace.
Amazon Web Services (AWS) announced the general availability of AWS Glue 6.0, delivering lower pricing than previous AWS Glue versions and introducing full support for Apache Iceberg v3 features. AWS Glue 6.0 is built on a fully modernized runtime, Apache Spark 4.1, Python 3.13, and Scala 2.13, delivering faster performance. With this release, AWS Glue provides a comprehensive Iceberg v3 implementation on any fully serverless managed Spark service, along with new capabilities that simplify ETL authoring, improve PySpark performance, and enable real-time streaming with single-digit millisecond latency.
BrainChip Holdings released the Symphony Community Akida Bundle, a free, open-source software bundle that lets developers run BrainChip Akida neuromorphic processors alongside their existing compute under a workload scheduler built on IBM Spectrum Symphony Community Edition. This unifies CPU, GPU, and neuromorphic workloads, enabling heterogeneous computing without compromising on performance and accuracy. Adding Akida to the IBM scheduling pool ensures that lightweight, event-driven inference can run rather than simply defaulting to a GPU.
Cloudera announced native GPU acceleration for Apache Spark 4.1 in Cloudera Data Engineering, enabled by the NVIDIA CUDA-X library, cuDF. The NVIDIA cuDF plug-in for Apache Spark will support the just-announced Cloudera Anywhere Cloud, which is designed to enable organizations to accelerate Spark workloads without rewriting PySpark or SQL code. This helps data teams prepare AI-ready data faster while reducing cloud infrastructure costs across hybrid environments.
The Cloud Native Computing Foundation (CNCF) announced the graduation of Kubeflow, a cloud native ecosystem forged by an open community dedicated to standardizing Data & AI workloads on Kubernetes. The graduation signals Kubeflow’s technical maturity and confirms its role as the operational backbone for enterprises running AI workloads in production, including data processing, model training, fine-tuning, and inference. Kubeflow provides native capabilities for data processing, interactive workloads, model training, fine-tuning, and interactive development. Kubeflow’s upcoming roadmap focuses on expanding Large Language Model (LLM) orchestration, enhancing post-training capabilities with fine-tuning, large-scale data engineering, and agentic workloads for the Data & AI lifecycle.
Concentric AI announced an industry-first vision model feature in the Concentric AI Semantic Intelligence platform that delivers an efficient and reliable way to detect sensitive documents that have unique visual signatures. The feature allows Concentric AI to analyze and understand the content of an image, such as the biographic (or photo) page of a passport from any country or a driver’s license from any state in the U.S., based on its visually consistent characteristics, even if the contents are blurred, in order to identify document type.
Databricks introduced Precision Mode in its document extraction API, ai_extract, to help teams extract structured data from complex enterprise documents. The solution is designed for complex documents that standard extraction tools often struggle with, including long PDFs, dense tables, charts, and information spread across multiple pages. Instead of relying on a single-pass extraction, Precision Mode uses custom-trained extraction models and an agentic harness to break complex jobs into smaller tasks, run them in parallel, and reconcile the results into a final structured output.
Eyelit Technologies introduced Agent EyeQ, an AI-driven application suite embedded directly within the manufacturing stack. Unlike traditional AI overlays, Agent EyeQ integrates seamlessly, enabling autonomous or semi-autonomous operations across scheduling, quality, and yield management. This integration allows the system to not only detect process excursions but also trace root causes, model downstream impacts, and prescribe corrective actions in a continuous reasoning loop. Every decision made by Agent EyeQ is logged, traceable, and fully explainable, ensuring compliance and transparency.
Globant introduced Glob.AI, a new AI-native tech services model that changes how enterprises access, purchase, and deploy AI services. Available online through a self-service model, Glob.AI gives organizations access to enterprise-grade, high-quality AI Pods (service units run by a set of AI agents and supervised by humans). Companies are charged for what they produce (per output or per consumption), never per seat or per hour.
Lucidity announced the extension of the Lucidity Platform into object storage. The company’s object storage support is built on the same visibility, recommendations, and actions framework that powers the rest of Lucidity. It starts with visibility and insights. Lucidity maps the data hierarchy across accounts, containers, folders, and prefixes to build a multi-layer heatmap of the entire environment. The heatmap surfaces size, cost, tags, tier, and access, modification, and creation age at every level, so teams can see which specific elements are driving cost in seconds instead of chasing them down by hand.
NeuBird AI announced the Earned Autonomy Framework, a set of open architectural principles defining how autonomous agents should earn, scope, and operate within write access permissions in production environments. The company is inviting operators, security leaders and engineers to co-sign it as a shared industry reference. The framework defines a four-level trust spectrum across which agents operate and progress based on verified performance, with permissions bounded by policy and revocable the moment confidence drops.
NTT DATA and Palo Alto Networks announced a multi-year strategic alliance designed to help organizations securely adopt AI, modernize cybersecurity, simplify complex technology environments and build cyber resilience for the AI era. The agreement combines Palo Alto Networks AI-powered cybersecurity platforms with NTT DATA’s consulting, engineering and managed services. Specifically, the alliance brings together Palo Alto Networks Unit 42 threat intelligence with NTT DATA’s global cybersecurity expertise, AI governance, and managed services.
Nutanix and ChronoScale Holdings Corporation announced a strategic partnership to jointly deliver enterprise-ready AI infrastructure and help accelerate adoption of AI services across global markets. The partnership brings together complementary capabilities enterprise customers have historically had to assemble themselves. The companies will combine Nutanix’s full portfolio of agentic AI solutions with ChronoScale’s accelerated compute, enterprise AI foundry, and outcome-driven delivery model.
Sauce Labs announced that bring-your-own-model capabilities are available now to Sauce Labs enterprise customers. Organizations can work with their Sauce Labs account team to enable model choice within AURA. To that end, AURA now supports open-source, open-weight, or proprietary large language models (LLMs) as its underlying model layer, standardizing release assurance without forcing organizations to rely on a single AI provider for coding.
Tricentis announced Tricentis Transform, a lineup of new AI-powered innovations designed to help enterprises build, validate, and deliver high-quality software at scale. The announcement includes three new technologies developed through Tricentis Labs, including: Tricentis Aida, an AI agent that autonomously explores web and Windows desktop applications, surfaces defects and coverage gaps, and delivers clear signals on application health without requiring an existing test suite, scripting, or setup. Tricentis AgentScore, a technology that shifts quality engineering from deterministic testing to probabilistic evaluation. And Tricentis Release Risk Intelligence, an intelligent capability that helps quality engineering leaders and release managers understand what’s at risk in a release and what actions to take next.
UiPath announced UiPath Maestro Flow, a developer-first orchestration canvas combining the speed of modern, AI-native development with enterprise-grade durability and governance. Using Maestro Flow, builders can use supported coding agents as a single artifact, from prototype to production. Specifically, builders can use the coding agents they already rely on, including Claude Code, Cursor, GitHub Copilot, and Codex, to design, run, observe, and govern complete business processes as a single artifact, all from their native development environment, such as VS Code or UiPath Studio.
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:
- Real-time Analytics News for the Week Ending August 15
- Real-time Analytics News for the Week Ending August 8
- Real-time Analytics News for the Week Ending August 1
- Real-time Analytics News for the Week Ending July 25
- Real-time Analytics News for the Week Ending July 18
- Real-time Analytics News for the Week Ending July 11
- Real-time Analytics News for the Week Ending July 4
- Real-time Analytics News for the Week Ending June 27
- Real-time Analytics News for the Week Ending June 20
- Real-time Analytics News for the Week Ending June 13
- Real-time Analytics News for the Week Ending June 6
- Real-time Analytics News for the Week Ending May 30
- Real-time Analytics News for the Week Ending May 23
