Israeli startup Hailo has extended its Series C with $120 mn to develop specialized chips for running generative AI tasks locally on so-called edge devices rather than sending information to a data center.
Delek Motors, Delek CEO Gil Agmon, businessman Alfred Akirov, VC firm OurCrowd, the Zisapel family and others led the extension.
The raise brings the round’s total to $256 mn and Hailo’s total capital raised to about $344 mn. The raise brings Hailo’s valuation to $1.2 bn, up from $965 mn as of 2021.
Hailo has made strides in the field of artificial intelligence, specifically through developing specialized chips for deep learning on edge devices.
Founded in 2017 by Orr Danon, Hadar Zeitlin, and Avi Baum, who all share backgrounds in the Israel Defense Forces’ elite technology unit, Hailo has positioned itself as a notable player in the tech industry, particularly in the AI and machine learning sectors.
As GenAI on the edge becomes immersive, the focus turns to handling large LLMs in the smallest possible power envelope. AI has to become fully immersive and always available
Orr Danon, Hailo CEO and co-founder
Hailo’s flagship product is the Hailo-8 Deep Learning Processor. This processor is designed to deliver high performance for AI applications at the edge, meaning it can run AI processes efficiently without needing to communicate with central servers.
The chip stands out for its ability to perform deep learning tasks with high energy efficiency, which is particularly valuable for devices that operate at the edge of the network where computing resources are limited.
Although Hailo’s technology is applicable in various industries, the company primarily targets sectors such as automotive, smart cities, and industry 4.0, where edge computing is crucial.
Their processors are advantageous for real-time applications that require immediate data processing without the delays involved in cloud computing.
By enabling smarter, faster edge computing, Hailo’s technology allows for more sophisticated and autonomous AI applications closer to where data is collected, reducing reliance on cloud-based infrastructure.
As the Internet of Things (IoT) and autonomous devices continue to evolve, the demand for high-performance, low-power AI processors like the Hailo-8 is expected to grow.
Hailo’s innovations position it well to take advantage of these trends, potentially making it a key enabler of the future AI-driven technology landscape.
The company’s ongoing development and scaling of its product offerings will be critical as it seeks to expand its presence and capitalize on global market opportunities.
Startups like Hailo—those creating application-specific chips—have weathered the VC downturn relatively well.
According to PitchBook data, there was a slight bump in 2023 funding and deal value relative to 2022, with $1.9 bn generated across 34 deals.
With all the excitement surrounding generative AI, Danon said the Series C extension was easier to raise.
Much of the VC investment surrounding generative AI has been directed at startups serving data centers that process the intensive workloads of large language models.
Lambda, a GPU computing startup, raised a $320 mn Series C in February, while Foundry raised $80 mn in March for machine-learning-specific cloud platforms.
But some tasks require immediate, low-latency processing. Enter edge devices: Lauded for being energy efficient, more secure and reliable, they handle applications at the source, not outsourcing requests to a data center, by processing data and running large language models like Meta’s Llama 2 locally.
Cars utilize edge devices for self-driving functionality, processing and translating data quickly when on the road. Investors hope startups like Hailo can bring generative AI to cars, industrial spaces and other sectors like medicine while lowering costs.
by Peter Sonner
