Myrtle.ai (Myrtle Software)

Myrtle.ai is a UK-based company that specializes in low-latency AI inference solutions for machine learning models. Founded in 2017 and headquartered in Cambridge, the company aims to make low-latency AI accessible by enhancing latency, throughput, energy efficiency, and cost across inference workloads. Myrtle.ai offers several specialized products, including VOLLO, an inference accelerator designed for capital markets that achieves microsecond-level latency. The SEAL Accelerator optimizes recommendation models and memory-intensive workloads, while the MAU Accelerator focuses on accelerating RNNs and deep neural networks for hyperscale inference. The company serves various applications, such as recommendation systems, speech processing, computer vision, natural language processing, finance, cybersecurity, and conversational AI. Myrtle.ai collaborates with leading technology partners like Altera, Achronix, and AMD to enhance its solutions.