Intel announced Hala Point, the world’s largest neuromorphic computer system, using hundreds of Intel’s powerful Loihi 2 parallel processors. It is designed to support brain-inspired artificial intelligence research and outperforms all existing systems. Hala Point is capable of 20 petaops and achieves over 15 trillion TOPS/W, exceeding all GPU and CPU built architectures. It supports various AI applications such as scientific problem-solving, language models, and AI agents. Hala Point packages 1,152 Loihi 2 processors in a single data center chassis with 1.15 billion neurons, 128 billion synapses, and over 2,300 embedded x86 processors, consuming 2,600 watts of power. It provides massive bandwidth for processing, memory, and communication, including a total of 16 petabytes per second (PB/s) of memory bandwidth, 3.5 PB/s of inter-core communication bandwidth, and 5 terabytes per second (TB/s) of inter-chip communication bandwidth.
The system’s full capacity of 1.15 billion neurons is 20 times faster than a human brain, and up to 200 times faster at lower neuron capacity. It also performs AI inference and optimization problems using 100 times less energy at speeds 50 times faster than conventional CPU and GPU architectures.
The whytry.ai article you just read is a brief synopsis; the original article can be found here: Read the Full Article…
