XCENA, a semiconductor startup, has successfully closed a $135 million Series B funding round, bringing its total capital raised to $185 million. The company, currently valued at $570 million, aims to address a critical inefficiency in artificial intelligence infrastructure: the data movement bottleneck. While much of the industry focus remains on processing power, XCENA is betting that the real challenge lies in how data travels between memory and processors.

The startup’s flagship product, the MX1 chip, is designed to bridge the gap between CPUs, GPUs, and DRAM. By utilizing Compute Express Link (CXL) technology and RISC-V based cores, the MX1 brings computational capabilities directly into the memory module. This approach minimizes the energy-intensive and time-consuming ’round trips’ that data currently makes during AI inference tasks. The company suggests that this architecture could potentially allow tasks that previously required ten servers to be handled by a single unit, offering massive efficiency gains for hyperscale data centers.

Founded in 2022 by industry veterans from Samsung and SK Hynix, XCENA is positioning itself against established players like Marvell and Astera Labs. Unlike competitors that may rely on general-purpose cores, XCENA utilizes a highly integrated design featuring thousands of small, efficient cores optimized for data orchestration. The company’s leadership believes the shift toward memory-centric architectures is inevitable as the demand for AI scaling continues to outpace traditional compute improvements.

The Series B round was co-led by Seoul-based venture capital firms Atinum and IMM Investment, with participation from Corstone Asia and existing backers including SBI Investment and Mirae Asset Capital. While the MX1 is currently in the prototype phase, XCENA plans to begin mass production through Samsung’s foundry lines by the end of 2026, with revenue generation expected to follow in 2027.