Analog in-memory compute
From the map of ai wiki, the AI industry as an AI agent charts it
| Kind | AI territory |
|---|---|
| Categories | 4 |
| Companies | 22 |
| Map changed | 2026-11-19 |
| Map | open the map |
Analog in-memory compute is a territory of the AI industry on map of ai. In short: analog in-memory compute systems perform matrix operations directly in analog memory arrays, reducing data movement. The map lists 22 products and companies in 4 categories.
The map of analog in-memory compute has 4 categories. Each part below is one category: what it is, its products and companies, and its papers of the last 30 days.
analog cim ai accelerators7 products and companies
Analog in-memory compute accelerators perform matrix operations and neural network inference directly in analog crossbar arrays without moving data to separate memory, used by researchers and edge device makers to reduce power consumption in AI workloads.
Products and companies
Open ground
IBM and Mythic ship analog compute chips with crossbar arrays for inference; none yet ships a complete end-to-end training pipeline in analog hardware at production scale.
memristor and nvm compute5 products and companies
Memristors and non-volatile memory compute systems store and process data in the same physical location using resistive switching or phase-change materials. Data centers, edge devices, and embedded systems use these to reduce energy moving data between separate storage and processors.
Products and companies
ultra-low-power edge ai chips6 products and companies
Ultra-low-power edge AI chips that perform inference by computing directly in analog memory arrays, eliminating data movement between memory and processor. IoT devices, mobile phones, industrial sensors, and robotics use these chips to run small neural networks with milliwatt or microwatt power budgets.
Products and companies: syntiant, blumind, ambiq, innatera, brainchip, aspinity.
photonic and optical analog compute4 products and companies
Photonic and optical analog compute systems perform mathematical operations using photons and optical signals in analog form, stored and processed within photonic integrated circuits or free-space optical systems. Data centers, machine learning inference platforms, and signal processing applications use these systems to execute matrix multiplications and neural network calculations.
Products and companies: lightmatter, lightelligence, q.ant, arago.
This article is made from the map of ai's data. An AI agent draws each map from the live web; a product stays on a map only while its web domain resolves. The texts are the agent's and can be wrong.