SAN JOSE, Calif. — December 2, 2025 — Leads & Copy — Astera Labs, Inc. (Nasdaq: ALAB) has announced plans to deliver custom connectivity solutions designed to address the growing complexity of next-generation AI infrastructure.
These solutions will optimize performance, energy efficiency, and cost for various workloads. Building on its May 2025 announcement of joining the NVIDIA NVLink Fusion Ecosystem, Astera Labs will collaborate with hyperscaler partners to develop custom solutions supporting NVLink connectivity.
As AI infrastructure becomes more complex, hyperscalers require tailored connectivity solutions to meet their architectural needs and performance goals, according to the company. Astera Labs’ custom connectivity solutions will be an extension to the company’s industry-standards-based Intelligent Connectivity Platform.
These custom solutions will enable infrastructure providers to design heterogeneous systems, integrating diverse accelerators into high-performance rack-scale platforms.
According to Sanjay Gajendra, President and COO at Astera Labs, hyperscalers need proven connectivity partners. The collaboration with NVIDIA enables greater flexibility for XPU connectivity through NVLink Fusion-enabled solutions, he said. With NVLink Fusion support, the custom connectivity solutions will be engineered to sustain multiple terabytes per second of low-latency data throughput and are expected to increase Astera Labs’ footprint in next-generation heterogeneous AI infrastructure.
The new custom connectivity solutions will leverage Astera Labs’ expertise in delivering purpose-built connectivity solutions and COSMOS software-defined architecture. Newly acquired capabilities, including photonic chiplet technology, are expected to enhance the company’s vision to provide a total offering for hyperscalers deploying standard, custom, and hybrid AI infrastructure.
Ashish Karandikar, Vice President of Engineering at NVIDIA, said that NVLink Fusion, combined with Astera Labs’ connectivity solutions, gives customers access to the industry’s best scale-up interconnect with NVLink in a rack-scale architecture. Karandikar added that Astera Labs joining the NVLink Fusion ecosystem gives innovators more choices in bringing semi-custom AI systems to market faster with higher bandwidth and lower latency performance.
Astera Labs provides rack-scale AI infrastructure through connectivity solutions. Its Intelligent Connectivity Platform integrates CXL®, Ethernet, NVLink, PCIe®, and UALink™ technologies with the COSMOS software suite to unify diverse components into systems that deliver end-to-end scale-up and scale-out connectivity. The custom connectivity solutions business complements its standards-based portfolio, enabling customers to deploy tailored architectures to meet their infrastructure requirements.
Cautionary Note Regarding Forward-Looking Statements: This communication contains forward-looking statements regarding Astera Lab’s NVLink Fusion-based customized connectivity solutions, the market’s future needs for such solutions, the anticipated impact and benefits of such solutions, and our expectations as to the collaborations relating to such solutions. Such statements involve risks and uncertainties, many of which are beyond the control of Astera Labs, our customers and our collaboration partners, that could cause actual results to differ materially from those expressed or implied in the forward-looking statements. Readers are cautioned not to put undue reliance on forward-looking statements, and no person assumes any obligation to update or revise any such forward-looking statements, whether as a result of new information, future events or otherwise, except to the extent that disclosure may be required by law.
Peter Lo
Peter.lo@asteralabs.com
Source: Astera Labs
