Overview of Open Edge Platform 2025.2: Suites, Libraries, Toolkits, and Frameworks to Propel Edge AI into 2026

[stevenhoenisch]

Open Edge Platform from Intel delivers a modular, composable stack of open-source software optimized to rapidly deploy, secure, manage, and scale AI solutions at distributed edge sites. To blaze an edge AI trail for the new year, we released Open Edge Platform 2025.2 in December.

The release lights up the first few months of 2026 with AI models, sample applications, SDKs, microservices, edge infrastructure, performance-optimized pipelines and workflows, AI-optimized code, benchmarking tools, and industry-specific libraries. The newly minted Robotics AI Suite takes flight, and the latest Geti release lets you fine tune vision models on Intel® Core™ Ultra Series 2 processors with a built-in GPU so you can find your own way with ease.

With the platform's suites and libraries, edge AI use cases span industries and fields like:

• Robotics
• Education
• Smart Cities
• Intelligent Retail Systems
• Manufacturing and Industrial IoT
• Streaming and Multimedia AI
• Multimodal Edge AI Applications
• Deep Learning, Computer Vision, and Generative AI

📑 Find out more by paging through our Medium blog post marking the release of Open Edge Platform 2025.2.

Platform Components

Open Edge Platform comprises Edge AI Suites, Edge AI Libraries, Edge Microvisor Toolkit, and Edge Manageability Framework. Vendors, developers, and technology partners can take part in the GitHub community for these solutions in various ways: contributing code, proposing a design, downloading and trying out releases, opening an issue, benchmarking application performance, and -- perhaps most important of all for building community -- participating in discussions.

For more information on Open Edge Platform, check out these resources and the sections later in this post on contributing to or participating in each component:

• Read and respond to blog posts
• Watch demo videos on YouTube

Edge AI Suites

Here's what’s new or enhanced for Edge AI Suites in the 2025.2 release:

• Robotics AI Suite goes gold with ROS 2 and GSML camera integration. A new model predictive control demo minimizes jitter and collision risks with a data-defined, linear action timing model. The new Robot Motion Control Task C++ library simplifies robot application development with integrated support for AI inference engines and EtherCAT protocol.

• Metro AI Suite: An Agentic AI Transit Route System tutorial uses simultaneous localization and mapping, real-time scene intelligence, and agentic AI for optimized, up-to-the-minute route finding.

• Retail AI Suite: An Order Accuracy sample pipeline enables video object detection across multiple cameras in real time, using a Vision Language Model to classify and validate quick-service food orders.

• Manufacturing AI Suite: A Multimodal Weld Detection pipeline processes visual data across multiple cameras and sensors with time-series and fusion analytics plus vision AI to find weld defects.

• Education AI Suite (Preview): A Smart Classroom sample application preview transforms audio and video from recorded class sessions into concise educational summaries on flat-panel displays using multi-modal AI models for audio transcription and video summarization.

💨 To find out about the many updates in this release of Edge AI Suites, check out the release notes on our documentation website.

Participate in or Contribute to Edge AI Libraries

Here's how to participate in or contribute to Edge AI Suites:

• Participate in discussions
• Open an issue
• Submit a pull request
• Read the Contribution Guide
• Report a security vulnerability
• Read and respond to blog posts
• Watch demo videos on YouTube
• View the documentation website
• See the latest weekly build report 🏗️

Edge AI Libraries

The 2025.2 release of Open Edge Platform and the Edge AI Libraries simplifies computer vision model fine tuning, speeds up video streams processing, and enhances provisioning for more streamlined and secure model deployment. In general, this release enables validation and support for Intel® Core™ Ultra Series 2 processors, utilizing built-in GPU and NPU to accelerate vision AI model training and fine-tuning. OpenVINO™ simplifies NPU deployment and enables secure Gen AI model deployment. And Geti™ introduces support for model fine-tuning on Intel® Core™ Ultra Processor Series 2.

This release also brings in broad camera support for multi-camera media streams and faster frames per second, and new sample applications using transformer-based models for speech recognition and real-time scene intelligence.

Here are some highlights of the release:

• Geti™: Enables model fine tuning on Intel® Core™ Ultra Series 2 processors with built-in GPU, putting more power into the hands of developers and enabling more continuous model fine tuning capabilities at the edge. Geti also adds a new state-of-the-art DEIM-DETR object detection model algorithm, which is especially helpful for small object detection use cases. Support for annotation shapes with holes enables model training for complex segmentation tasks for manufacturing and medical use cases. And further flexibility on data augmentation enables developers to fine-tune models by configuring selective data augmentation parameters.

• SceneScape: The scene intelligence microservice adds vertical scaling to track up to a hundred objects through time chunked processing. It also added a new object clustering service to analyze object properties as well a new mapping service to create 3d meshes and automatically calibrate cameras from camera inputs. These services simplify developer workflows for calibration as well as enabling scaling many object tracking to build a more comprehensive 3d scene intelligence application.

• OpenVINO™ toolkit: OpenVINO adds in more new model support, including a preview of Mixture-of-Experts (MoE) models, such as Qwen 3, optimized for CPUs and GPUs, as well as an encrypted blob format for secure deployment of OpenVINO GenAI models to help ensure that IP thefts for model weights and artifacts can be avoided in deployment.

• DL Streamer: Smart video processing, auto-optimizing camera settings, and unique AI-powered fingerprinting of objects for re-identification enables developers to enhance their video analytics application development workflows.

In addition, the Edge Workloads and Benchmarks Pipelines repository taps performance-optimized pipelines utilizingthe GStreamer multimedia framework and the Deep Learning Streamer (DL Streamer) to measure and validate media and edge AI video analytics.

📣 Get the full story on all the enhancements in the release notes.

Participate in or Contribute to Edge AI Libraries

Here's how to participate in or contribute to Edge AI Libraries:

• Participate in discussions
• Open an issue
• Submit a pull request
• Read the Contribution Guide
• Report a security vulnerability
• Read and respond to blog posts
• View the documentation website
• Read the release notes

🏗️ To download a library, see the weekly build reports.

Edge Microvisor Toolkit

The latest release of Edge Microvisor Toolkit helps builders and devops engineers assess edge AI deployment scenarios on Intel processors and to ensure that they can take advantage of the latest hardware capabilities.

As a reference Linux operating system, Edge Microvisor Toolkit demonstrates the full capabilities of Intel processors for Edge AI workloads through Linux patches from Intel that are yet to be upstreamed, and you can, for instance, see some of those Linux patches for GPU and NPU support in the toolkit's SPEC file.

• Check out the release announcement or the release notes to see the changes, enhancements, and the latest CVE fixes.

Another way to deploy a version of Edge Microvisor Toolkit for a benchmarking effort is to use the newly released OS Image Composer and one of its many image templates, including templates for generating versions for use with another component of Open Edge Platform: Edge Manageability Framework. Find out more at https://github.com/open-edge-platform/os-image-composer.

• Get the latest weekly release of the RAW image for the real-time or non-RT Developer Node. Its latest weekly release is posted in GitHub Discussions under 📣 announcements.

• Build the immutable Standalone Node.

• Build your own custom mutable or immutable version by selecting a predefined JSON image configuration file and a package list file that determine the image's properties and capabilities. In addition, you can create your own image configuration file and package list to build exactly the version that you want -- a powerful approach to customizing a Linux operating system to validate your edge AI workload on Intel processors.

For help installing and configuring a version of Edge Microvisor Toolkit, see the the documentation website.

Participate in or Contribute to Edge Microvisor Toolkit

Here are other ways to participate in or contribute to Edge Microvisor Toolkit:

• Participate in discussions
• Open an issue
• Submit a pull request
• Read the Contribution Guide
• Submit a design proposal
• Report a security vulnerability
• Read and respond to blog posts
• View the documentation website
• Download the latest weekly release 🏗️

Edge Manageability Framework

Release 2025.2 of Edge Manageability Framework helps system builders and DevOps engineers assess deployment scenarios on Intel processors to ensure that they can take advantage of the latest hardware capabilities as well as the remote management capabilities powered by Intel vPro Essentials and vPro Enterprise.

For the first time, this release lets you selectively omit layers of the stack to reduce resource requirements. At the same time, this release improves support for operating systems, including the provisioning of Ubuntu 24.04 -- improvements that complement the release of another tool that loosely falls under the Open Edge Platform rubric: OS Image Composer.

Similarly, this EMF release adds out-of-band management support for Intel Core platforms featuring Intel® vPro® Essentials Platform in addition to the existing support for Intel vPro® Enterprise Platform. There is support for using Intel vPro platform-based out-of-band management for Intel Standard Manageability in conjunction with Intel Active Management Technology (Intel® AMT) devices.

📣 Line items in the limelight: Here are a few highlights of the 2025.2 release:

• Alpha support for Intel® Core™ Ultra Series 3 processors (formerly code-named Panther Lake) supporting integrated GPU and NPU.
• Support for Ubuntu 24.04 LTS OS with kernel 6.11.0-17. Official support for Ubuntu 22.04 LTS OS continues.
• Ability to create custom OS profiles based on Ubuntu 24.04 LTS or Edge Microvisor Toolkit.
• Customization capabilities for the operating system kernel command-line parameters.
• Expanded support for reference applications like Pallet Defect Detection, Image-Based Video Search, Chat Q&A Sample Application, and Smart Parking.
• Security compliance verification through tracking of open and fixed CVEs for packages installed on Ubuntu 24.04 LTS OS and Edge Microvisor Toolkit.

💨 Fast and fluent: Check out the version 2025.2 key performance indicators, or KPIs, that demonstrate quality and proficiency for edge AI workloads, including Open Edge Platform applications like smart parking, defect detection, and image-based video search.

🚀 Image and substance: With the newly released OS Image Composer, you can use prefabricated YAML image templates of Edge Microvisor Toolkit and its real-time variant to generate versions for use with Edge Manageability Framework. Find out more at https://github.com/open-edge-platform/os-image-composer.

🎉 New and updated: So why not ring in the holidays by checking all the new features and capabilities as well as updates to existing capabilities for this 2025.2 release of Edge Manageability Framework listed in the release notes and let us know your thoughts in the GitHub Discussions?

Participate in or Contribute to Edge Manageability Framework

• Participate in discussions
• Open an issue
• Submit a pull request
• Submit a design proposal
• Report a security vulnerability
• Read the latest release notes
• Read and respond to blog posts
• View the documentation website
• See the latest weekly builds 🏗️

---