Announced details
Metis AIPU maximum processing performance at INT8: 214 TOPS
Power efficiency: 15 TOPS/W
Marubun Corporation signed a VAR agreement with Axelera AI B.V. Marubun Corporation will provide the products with added value, including technical support.
Based in Eindhoven, the Netherlands, and established in 2021. Its representative is Fabrizio Del Maffeo (CEO & Co-Founder). Axelera AI B.V. is an AI semiconductor manufacturer that develops, manufactures and sells M.2- and PCIe-form-factor modules and board products equipped with Metis.
In this announcement the company addresses none of Contract Value / Offering price / Impact on Business Performance.
| Item | Details |
|---|---|
| Names of Products / Services Covered | M.2 module equipped with a Metis AIPU, PCIe expansion card equipped with a Metis AIPU, Voyager SDK, Development single-board computer (SBC) |
| Counterparty | Axelera AI B.V. |
| Counterparty overview | Based in Eindhoven, the Netherlands, and established in 2021. Its representative is Fabrizio Del Maffeo (CEO & Co-Founder). Axelera AI B.V. is an AI semiconductor manufacturer that develops, manufactures and sells M.2- and PCIe-form-factor modules and board products equipped with Metis. |
| Relationship with the Company | Marubun Corporation signed a VAR agreement with Axelera AI B.V. Marubun Corporation will provide the products with added value, including technical support. |
| Contract Date | 19 August 2026 (the announcement date of the agreement; the agreement date itself is Not disclosed) |
| Applicable area | Security, Industrial automation, Smart cities, Retail, Healthcare, Agriculture, Robotics, Social Infrastructure, Manufacturing |
| Quantitative Effects | Metis AIPU maximum processing performance at INT8: 214 TOPS, Power efficiency: 15 TOPS/W |
Items this document does not state (11)
These are items the company does not state in this document. We do not leave them blank, nor fill them in from other sources or by inference.
The reasons the company gives
The company's own explanation, not something inferred from the numbers.To support customers in streamlining product development, accelerating time to market, and reducing the total cost of ownership when implementing edge AI, amid growing demand for high-performance real-time processing in the edge AI field.