SOURCE-LINKED INTELLIGENCE
Scalable Integrated Reconfigurable Energy-Efficient Neuromorphic Accelerators
Scalable Integrated Reconfigurable Energy-Efficient Neuromorphic Accelerators The explosion of artificial intelligence (AI) has unleashed a gold rush for AI accelerator chips, specialized chips that power everything from chatbots to self-driving cars. The large energy footprint forms a major global problem. SIRENA is Europe’s leap to build a new semiconductor business that makes AI efficient from edge to cloud. We will industrialize Reconfigurable Nonlinear-Processing Units: CMOS-compatible devices that perform analogue signal transformation and feature extraction with tunable nonlinearity and local adaptivity. The result is 10x lower energy, reduced latency, and far higher density than today’s digital accelerators. RNPUs complement CPUs, GPUs, NPUs, or power standalone accelerators, enabling always-on intelligence in watches, cars, sensors, and servers. Backed by EIC Pathfinder HYBRAIN results and a recent Nature study, SIRENA lifts the platform from proof of principle to TRL 6
Read original source ↗ Open in workspace
- recordType
- award
- status
- SIGNED
- region
- EU
- value
- 2498969.75
- unit
- EUR
Evidence & attribution
European Commission, CORDIS Horizon Europe project dataset. Metadata adapted.
License: CORDIS reuse policy
First collected: 2026-09-20T05:31:32.981Z. This is not the publication date.