The new Pulsar chip brings brain-inspired intelligence to battery-powered devices for real-time, ultra-low power AI at the edge.
Innatera, a developer of neuromorphic processors, recently launched Pulsar, its first commercially available microcontroller designed to bring brain-like intelligence into edge devices. Neuromorphic processors are computing chips designed to mimic the structure and function of biological neural networks, particularly the human brain. Unlike traditional digital processors that use the von Neumann architecture (separate memory and processing units), neuromorphic chips integrate memory and computation in the same physical locations, similar to how neurons and synapses work together in the brain.
Born from more than a decade of research, Pulsar processes data locally at the sensor level, eliminating the need to rely on brute-force compute in power-hungry edge processors or data centers to make sense of sensor data. Innatera claims that it also delivers up to 100 times lower latency and 500 times lower energy consumption than conventional AI processors.
The new microcontroller introduces a compute architecture based on Spiking Neural Networks (SNNs), a generational leap in AI hardware that processes data the way the brain does, focusing only on changes in input. This event-driven model reduces energy use and latency while delivering precise, real-time decision-making. Pulsar also combines neuromorphic compute with traditional signal processing and provides versatility by integrating a high-performance RISC-V CPU and dedicated accelerators for Convolutional Neural Networks (CNNs) and Fast Fourier Transform (FFT) in a single chip.
“By using brain-inspired Spiking Neural Networks, it brings real-time processing to ultra-low-power devices without leaning on the cloud. That means sensors that can think for themselves — faster responses, lower energy use, and smarter performance across everything from wearables to industrial systems,” said David Harold, senior analyst at Jon Peddie Research, in a press release.
With Pulsar, Innateria aims to give product teams a shortcut to smarter features that were previously off-limits due to size, power, or complexity. Filtering and interpreting sensor data locally keeps the main application processor asleep until truly needed, in some cases, eliminating the need for a main application processor or cloud computing, extending battery life by orders of magnitude. With sub-milliwatt power consumption, Pulsar makes always-on intelligence viable, enabling everything from sub-millisecond gesture recognition in wearables to energy-efficient object detection in smart home systems. For example, it can achieve real-time responsiveness with power budgets as low as 600 µW for radar-based presence detection and 400 µW for audio scene classification.
Innatera also aims to transform traditional sensors into self-contained intelligent systems. With its small memory footprint and efficient neural models, it fits into tight form factors while eliminating the need for heavy external compute and reducing reliance on complex, custom DSP pipelines. The idea is that sensor manufacturers can deliver plug-and-play smart sensor modules that accelerate development and time to market.
Using the company’s Talamo SDK, developers can build spiking models from scratch in a PyTorch-based environment and simulate, optimize, and deploy. Innatera is also launching a developer program, now open to early adopters, to provide a foundation for a growing community that accelerates innovation, shares knowledge, and empowers members to build the next generation of intelligent edge applications together. An upcoming open-source PyTorch frontend and marketplace will create an even more collaborative ecosystem for neuromorphic AI.
For more information, visit innatera.com/pulsar.
