Catalog of SIP Cores
System on Chip design resources
System on Chip design resources
June 9, 2026 -
Sophia Antipolis - Bringing together leading research institutions, semiconductor actors, and industrial partners across Europe, NimbleAI addressed a critical challenge: how to enable real-time, energy-efficient, and secure AI processing directly at the edge — where data is generated.
As AI systems increasingly move from cloud to real-world environments, the ability to process data locally has become a strategic necessity. This shift is particularly important for Europe, where control over critical technologies, data, and infrastructure is at the heart of current industrial and policy priorities.
From sensing to system integration: a holistic approach
NimbleAI delivered concrete results across the full technology stack.
The project developed a new generation of event-based vision sensors, enabling ultra-low latency perception while significantly reducing data processing requirements compared to conventional image sensors.
On the computing side, the consortium designed advanced AI processing architectures combining near-memory computing, RISC-V processors, and FPGA-based acceleration. These innovations address one of the key bottlenecks in modern AI systems — data movement — enabling more efficient execution of complex workloads under strict power constraints.
At system level, NimbleAI introduced adaptive and secure architectures capable of dynamic reconfiguration, enhanced monitoring, and safe remote updates — key capabilities for deploying AI systems in industrial and safety-critical environments.
A contribution to Europe’s technological sovereignty
Beyond technical achievements, NimbleAI contributes to strengthening Europe’s position in strategic technology domains, including edge computing, embedded AI, and advanced sensing.
By developing key building blocks within Europe — from hardware architectures to system integration — the project supports greater autonomy in the design and deployment of next-generation intelligent systems.
This aligns with broader European initiatives aimed at reducing dependency on non-European technologies and fostering a competitive, resilient semiconductor and AI ecosystem.
From research to real-world applications
The technologies developed within NimbleAI have clear application potential across multiple sectors, including robotics, industrial automation, space systems, and edge intelligence.
Through its dissemination activities, engagement with industrial stakeholders, and structured exploitation strategy, the project has laid the groundwork for future adoption and continued development beyond its lifetime.
A collective European achievement
NimbleAI demonstrates the value of collaborative European research in addressing complex technological challenges. By bringing together complementary expertise across disciplines and sectors, the consortium successfully delivered integrated solutions that would not have been possible in isolation.
As Europe continues to invest in strategic technologies, projects like NimbleAI highlight the importance of coordination, scale, and long-term vision in turning research into impact.
While the project formally concludes, its results are expected to contribute to ongoing innovation efforts and future developments in edge AI systems.
About NimbleAI
NimbleAI will design EDA tools to customize and integrate technologies and components (light-field-enabled dynamic vision sensing, event-based inference and processing, specialized processing with in-memory computing and programmable logic (eFPGA), embedded ReRAM-based storage, 3D integration of circuit layers (TSV-based inter-layer data movement), mutual adaptation of sensing and processing to operate at optimal DVFS point, dedicated software tools) and will deliver a prototype FPGA implementation of the 3D silicon stacked sensing and processing architecture, as well as the corresponding programming tools to develop and run computer vision applications on this architecture. The prototype will serve as a research vehicle for testing new concepts, new computer vision algorithms, and new runtime optimizations. The results of NimbleAI will lead to practical implementations in the next generations of commercially available neuromorphic chips. In fact, the project will fabricate some test chips to validate some key concepts in silicon.
The consortium is formed by the following organizations: Ikerlan S. Coop. (ES), Barcelona Supercomputing Center (ES), Menta SAS (FR), Universiteit Leiden (NL), Codasip s.r.o. (CZ), GrAI Matter Labs B.V. (NL), University Of Manchester (UK), Consejo Superior De Investigaciones Científicas (ES), Universitat Politècnica De València (ES), Monozukuri S.P.A. (ITA), Politecnico Di Milano (IT), Commisssariat A L'energie Atomique Et Aux Energies Alternatives (FR), IMEC - Interuniversity Microelectronics Centre (BE), Raytrix Gmbh (DE), AVL List Gmbh (AT), ULMA Medical Technologies S. Coop. (ES), Viewpointsystem GmbH (AT), Queen Mary University Of London (UK) et Technische Universitat Wien (AT).
NimbleAI has received funding from the EU’s Horizon Europe Research and Innovation programme (Grant Agreement 101070679), and by the UK Research and Innovation (UKRI) under the UK government’s Horizon Europe funding guarantee (Grant Agreement 10039070).