SiPearl CPUs will power IT4LIA supercomputer
April 22, 2026
SiPearl CPUs, paired with Axelera AI accelerators, will power the inference partition of IT4LIA.
Our solution
AI factories are purpose-built infrastructures that combine GPU clusters, high-speed interconnects and CPU orchestration layers to train and serve next-generation AI models.
The CPU plays a critical role: coordinating accelerator utilization, managing data pipelines, and handling the control plane logic that keeps the entire system running efficiently.
Typical workloads
-
Large language model pre-training
-
High-throughput AI inference
-
Data ingestion and preprocessing pipelines
-
Multimodal model training
-
Reinforcement learning from human feedback
Why SiPearl
→ High-bandwidth CPU architecture built to keep accelerators fully fed and utilised.
→ Energy-efficient design that lowers the cost per AI training step at scale.
→ European technology enabling sovereign AI capabilities outside US and Chinese supply chains.
Technical features of Rhea1
Discover all the details of our CPU
Seine Reference Server, a modular solution dedicated to Rhea1
Multifunctional, flexible and versatile, the Seine Reference Server can be used for validation and testing, as a reference design, for software porting and for demonstrations and customer testing.
It is available in two configurations:
- a single Rhea1 processor connecting to up to two GPUs in a single chassis,
- or as a dual socket Rhea1.
Each configuration supports up to two SATA (Serial Advanced Technology Attachment) disks and up to two PCIe NICs.
Our involvement in AI Factories projects
SiPearl, associate member of AION for AI gigafactories
May 20, 2026
The AION consortium has been formed to put forward an ambitious French bid under the European AI gigafactory programme.