In a significant advancement for edge computing, Hailo, a trailblazer in edge AI processors, has successfully completed a new funding milestone of $120 million, propelling its total investment to over $340 million. This financial boost coincides with the debut of its groundbreaking Hailo-10 GenAI accelerator, marking a pivotal shift towards embedding generative AI directly into edge devices across the personal computer and automotive industries.
The new funding round was led by current and new investors including the Zisapel family, Gil Agmon, Delek Motors, Alfred Akirov, DCLBA, Vasuki, OurCrowd, Talcar, Comasco, Automotive Equipment (AEV), and Poalim Equity.
Hailo CEO and Co-Founder Orr Danon stated, “The closing of our new funding round enables us to leverage all the exciting opportunities in our pipeline, while setting the stage for our long-term future growth. Together with the introduction of our Hailo-10 GenAI accelerator, it strategically positions us to bring classic and generative AI to edge devices in ways that will significantly expand the reach and impact of this remarkable new technology. We designed Hailo-10 to seamlessly integrate GenAI capabilities into users’ daily lives, freeing users from cloud network constraints. This empowers them to utilize chatbots, copilots, and other emerging content generation tools with unparalleled flexibility and immediacy, enhancing productivity and enriching lives.”
Founded in 2017 in Israel, Hailo has rapidly ascended to a prominent position in the AI chip industry, serving over 300 customers globally. With offices spanning the United States to Asia, Hailo’s innovative processors deliver data center-level AI performance on edge devices. These processors, redefining conventional computer architecture, facilitate real-time deep learning tasks with unmatched efficiency.
Hailo-10 Accelerator
The introduction of the Hailo-10 accelerator is set to redefine the user experience by enabling the execution of GenAI applications locally, without the dependency on cloud-based services. Orr Danon, Hailo’s CEO and Co-Founder, emphasizes the transformative potential of Hailo-10, highlighting its ability to integrate GenAI capabilities seamlessly into daily life. The Hailo-10 not only promises enhanced productivity and enriched user experiences but also addresses privacy concerns by processing data locally and reducing reliance on cloud data centers.
Hailo-10 stands out for its performance-to-cost and performance-to-power consumption ratios, supporting a wide range of applications while ensuring sustainability. Its compatibility with Hailo’s comprehensive software suite, shared with the Hailo-8 and Hailo-15 processors, facilitates seamless integration across various devices and platforms.
The Hailo-10 sets a new benchmark for edge AI accelerators with its capability to perform up to 40 tera operations per second (TOPS), outpacing integrated neural processing units (NPUs) in both speed and energy efficiency. Recent benchmarks reveal that it offers double the performance while consuming only half the power compared to Intel’s Core Ultra NPU.
Orr Danon further added, “Whether users employ GenAI to automate real-time translation or summarization services, generate software code, or images and videos from text prompts, Hailo-10 lets them do it directly on their PCs or other edge systems, without straining the CPU or draining the battery.”
Notably, Hailo-10 showcases remarkable efficiency, capable of running complex models like Llama2-7B and Stable Diffusion 2.1 within an ultra-low power envelope, setting new benchmarks in the field. The accelerator’s early applications are poised to revolutionize PCs and automotive infotainment systems by powering sophisticated AI features such as chatbots and personal assistants, heralding a new era of intelligent edge computing.
With sample shipments of the Hailo-10 GenAI accelerator anticipated in Q2 of 2024, Hailo is at the forefront of ushering generative AI into the mainstream, transforming how users interact with technology on a fundamental level.
Credit: Source link
