Proposals

Eclipse PanEval is an open-source large model evaluation platform and framework, designed to establish scientific, impartial, and open evaluation benchmarks, methodologies, and toolsets. It comprehensively assesses foundation model performance across language, multimodal, vision, and speech domains.

Core framework: A three-dimensional evaluation system based on "Capacity – Task – Metrics":
- Capacity: defines the scope of model capabilities ("What to evaluate?")
- Task: the form used to assess model capabilities ("How to evaluate?")
- Metrics: quantitative assessment from multiple perspectives ("How to measure?")

Eclipse PanEval covers 4 major model categories and 40+ evaluation tasks, with Safety & Robustness as a cross-cutting evaluation dimension for all categories.

Eclipse SDV Hephaestus ensures low efforts for contributors to participate in a Eclipse SDV project, the implementation is based on open source tools. The starting point for this project is the Eclipse S-CORE toolchain, which will be reviewed to determine how it can be made ready for use by other projects within Eclipse SDV Working Group. To ensure an qualification process compliant to ASPICE and ISO 26262, a solution will be developed to release a qualified product. This includes checking existing available approaches, such as those from RedHat.

Core Principles

• DRY (Don't Repeat Yourself): Centralize non-differentiating toolchain components so each project can focus on its domain-specific innovation
• Open Source by Default: Use open-source tools. Closed-source extensions may be integrated via well-defined plugin interfaces, but are never required
• Composable and Overlay Architecture: Projects can adopt the full toolchain or only the parts they need. Partial adoption is a first-class use case

The project will be hosted in one GitHub organization with multiple repositories. It will feature a modular architecture, ensuring that partial usage of the toolchain is possible. It will provide a central starting point for developers and support multiple languages, first including C++ and Rust.

Requirements will be defined through feature requests and issues, then reviewed with Eclipse SDV Working Group projects (e.g. presentation to the TAC).

Technical Details

The project is structured into four architectural layers, with AI integration running as cross-cuttingly across them.

Layer 1 - Build & Dependency Management:

Defines how the software is built and how dependencies are resolved. Built on Bazel as a starting point. 

Layer 2 - Environment Management & Tool Provisioning:

Ensures all required tools and compilers are available in the correct version and configuration. Standardizes Environment Setup with tools like OCI-Container and other tools e.j. Nix, Moon/Proto (to be evaluated).

Layer 3 - Task Runner & Automation:

Provides a uniform command line interface for common workflows e.g. (build, test, lint, format, flash, run, package, simulate). Shields developers from underlying tool complexity. Aligns CI/CD pipelines, build agents, and remote execution with the same conventions used locally.

Layer 4 - Development Environment & Enablement:

Delivers a ready-to-use, pre-configured development environment with no manual setup.

Cross-cutting AI Integration:
Provides intelligent support across all layers: analyzes build/test/tooling outputs, suggests fixes and improvements, assists with configuration and workflow execution. Initial use case: automated dependency tree reasoning explaining why direct and transitive dependencies are included, useful for FOSS analysis.

Pros and Cons of the Proposed Solution

Pros:

• Eliminates redundant toolchain development across SDV projects, freeing resources for domain innovation
• The harmonized Bazel-based build system simplifies multi-project integration significantly
• Modular design allows partial adoption, no all-or-nothing migration required
• Qualification readiness (ASPICE, ISO 26262) reduces compliance effort per project
• Automated SBOM generation and FOSS/SCA scanning reduces security and license review overhead

Cons:

• Governance complexity: cross-project decisions require broader consensus, which may slow iteration
• Migration effort: existing projects face upfront investment to adapt to the unified toolchain
• Dependency risk: adopters become coupled to the project release cadence and maintenance health

Eclipse Lyra is a modular, extensible web application framework for building desktop-like applications. It provides:

• Application and extension loader: Dynamic app registration and lifecycle, extension discovery and dependency resolution, and contribution registration.
• Contribution system: Declarative UI contributions (tabs, toolbars, commands, panes) targeting well-defined areas (e.g. sidebars, editor area, bottom panel).
• Command system: Context-aware commands with multiple handlers, key bindings, and exposure as tools for AI agents.
• Core services: Workspace (file system abstraction and persistence), settings, editor registry, task service, event bus, and dependency injection.
• Optional extensions: AI system (multi-provider LLMs, agents, tools), RAG, Monaco editor, notebook, Python terminal, Git, WebDAV, and others, loadable on demand.

The framework is built with TypeScript and Lit (web components standard). It is suitable as the base for desktop-style applications, coding environments, and other tooling that need a consistent UX model and extension story.

CORE-ETSP combines many-core RISC-V-based RTL with MRAM and thus creating a basis for the next generation ET Silicon Platform design. It can be deployed either in a traditional configuration with the host CPU accessing ETSP as an Intelligent RAM (replacing SRAM and Flash) via Hyperbus OR as a self-hosted array of microcontrollers (with or without a host CPU). 

When combined with the development platform (composed of various open upstream components), ETSP platform is a comprehensive solution for fast and low-power AI inference workloads at the edge. Multiple verticals and embedded AI systems in manufacturing, robotics and drones, and security systems may benefit from ETSP.

The Eclipse SDV Landscape project provides a comprehensive, visual, and continuously maintained overview of all Software-Defined Vehicle (SDV)–related projects hosted under the Eclipse Foundation. The project delivers a website that categorizes and presents Eclipse SDV projects in a clear, structured, and easily navigable way, supporting discover-ability, orientation, and communication across the SDV ecosystem.

The project builds on existing information of the Eclipse Foundation PMI and relies on existing official Eclipse Project APIs.

The Eclipse SDV Landscape is the official, community-maintained reference for the SDV Working Group and its stakeholders.

The project is providing 

1. Eclipse SDV Landscape public Website
   • Interactive visual overview of SDV projects
   • Hosted via Eclipse infrastructure or approved hosting
   • Organize projects by functional categories (e.g., tooling, middleware, RTOS, AI, digital twin, orchestration, etc.)
   • Exportable views (e.g., PNG, SVG) for presentations and documentation
   • Allow easy updates as projects are added, retired, or reclassified
2. Open Source Repository
   • Source code for generating and maintaining the landscape
   • Configuration and metadata describing SDV projects
   • Documentation for contributors and maintainers
3. Contribution Guidelines
   • Clear process for adding, updating, or removing projects
   • Alignment with Eclipse Foundation branding and policies

Out of Scope

The project will not:

• Replace official project documentation or websites
• Provide technical integration or dependencies between projects
• Act as a governance or compliance tool
• Host or mirror project artifacts or code beyond the landscape itself

The Eclipse Connector Fabric Manager includes:

- A tenant management and provisioning system that can be extended to support a variety of dataspace-related services.
 - A multi-role user interface that supports operators, B2B resellers, and end users for tenant management and service provisioning.
- A demonstrator sample

The Eclipse Azimuth Sailing Analytics, formerly known as the "SAP Sailing Analytics," offers a solution for portraying and analyzing sailing regattas, supporting training scenarios, and powering the vast archive at https://sapsailing.com. The solution consists of a cloud application with a web-based user interface, as well as three companion apps that integrate with the cloud application. The "sailing-analytics" repository has the code for the cloud-based web application, and two of the three mobile apps (Buoy Pinger and Race Manager). The third companion app (Sail Insight) is found in the "sailing-analytics-sail-insight" repo.

Eclipse VOStack is an open source software stack for IoT virtualization and convergence with edge/cloud computing technologies. It is aligned with the W3C Web of Things Standard.

The core of Eclipse VOStack is a python implementation building on wot-py (asynchronous Python implementation of a W3C Web of Things runtime). It extends wot-py with features of Virtual Objects (VO): 

• Extension of Protocol Bindings: HTTP(S), CoAP, WebSockets, MQTT, Zenoh
• Periodic Functions: Repeat Functions in irregular intervals
• InfluxDB: Automatic saving of Property values on read/write operations, logging of Action invocations and Event emission
• Automatic configuration of the VO: Script runner that takes a Web of Things Thing Description, a VO descriptor and a python code file and configures the VO

Additionally, VO Stack provides the following functionalities:

• Automatic orchestration/virtualization in Kubernetes deployments
• RTSP server for video streams
• Proxy mode to turn a cVO into a proxy to other VOs

The project provides an API to facilitate the execution of AI agents on Jakarta EE runtimes. It defines the API contract between the runtime and agent implementation. This is similar to what Servlet did for HTTP processing, Jakarta REST did for RESTful web services, or perhaps most appropriately, Jakarta Batch did for batch processing. The annotation-based API will define common usage patterns and life cycles that make it as understandable, consistent, and easy as possible to implement different types of agents. CDI is used as a core component model for the API. The project will likely also define integrations with other key Jakarta EE APIs such as Validation, REST, JSON Binding, Persistence, Data, Transactions, NoSQL, Concurrency, Security, Messaging, and so on. The project aims for the broadest industry consensus possible by engaging as many relevant subject matter experts and API consumers as possible, from within the Java/Jakarta EE ecosystem as well as externally.

Industrial and academic projects requiring access to either 32- or 64-bit, embedded or application class cores. The CVA6 can be synthesized for both FPGA and ASIC targets.