Axelera AI
Axelera AI accelerates edge AI inference with energy-efficient purpose-built hardware.
Axelera AI develops purpose-built hardware and software platforms that accelerate AI inference at the edge using proprietary digital in-memory computing (D-IMC) technology. Their flagship Metis™ AI Platform combines CPU and memory into a single chip, eliminating energy-intensive data movement between cloud and device, delivering 10x better energy efficiency than standard computing. The company serves industrial, defense, robotics, and retail customers who need low-latency, privacy-preserving AI processing on devices rather than cloud.
Problem solved
Organizations struggle with cloud latency, data privacy concerns, and mounting compute costs when processing AI workloads remotely instead of on edge devices.
Target customer
Industrial manufacturers, defense and public safety agencies, robotics companies, retail enterprises, and agritech firms deploying AI inference on edge devices where latency, privacy, and energy efficiency are critical.
Founders
F
Fabrizio Del Maffeo
CEO & Co-Founder
Previously Head of AI at Bitfury Group and VP/Managing Director at AAEON Technology Europe (ASUS Group).
E
Evangelos Eleftheriou
CTO & Co-Founder
IEEE Fellow and IBM Fellow with 35+ years at IBM Research Zurich; authored 250+ publications and holds 160+ patents in computing and memory technology.
M
Marco Barbato
Director of AI Integrated Systems & Co-Founder
20+ years in AI and Industrial IoT; previously AI & IoT Product Development Director at AAEON Europe and Director of AI Integrated Systems at Bitfury.
B
Bram Ernst Verhoef
Co-Founder, Head of AI
I
Ioannis Papistas
Co-Founder & Senior R&D Engineer
Funding history
Seed
$12M
2021
Led by Unknown
· Unknown
Series A
$27M
October 2022
Led by Innovation Industries
· Dutch government innovation credit (~$6.7M)
Series A Extension
$23M
2023
Led by Unknown
· Unknown
Series B
$68M
June 27, 2024
Led by European Innovation Council Fund
· Innovation Industries, Invest-NL, Samsung Catalyst Fund, Verve Ventures, Bitfury, SFPIM
EU Grant
€61.6M
March 2025
Led by EuroHPC Joint Undertaking
· DARE (Digital Autonomy with RISC-V for Europe) project
Series C/Growth
$250M
February 24, 2026
Led by Innovation Industries
· BlackRock, SiteGround Capital, Bitfury, CDP Venture Capital, European Innovation Council Fund, SFPIM, Invest-NL, Samsung Catalyst Fund, Verve Investments
Total raised:
$450M+
Pricing
Not publicly available. Metis PCIe and M.2 cards available through partner store, but pricing not listed. Enterprise/custom pricing model through direct sales and distribution partners.
Notable customers
500+ global customers across defense, public safety, industrial manufacturing, retail, agritech, robotics, and security sectors. Specific named customers not disclosed.
Integrations
Arduino, Aetina, Astute, C&T Solution, Eurocomposant, Macnica ATD Europe, Micron, Rutronik, Seco, Silicon Applications Group Corp (Partner Accelerator Network)
Tech stack
jQuery Migrate (JavaScript libraries)
jQuery UI (JavaScript libraries)
jQuery (JavaScript libraries)
core-js (JavaScript libraries)
Preact (JavaScript libraries)
Bootstrap Icons (Font scripts)
MySQL (Databases)
WordPress (Blogs)
MonsterInsights (Analytics)
Quantcast Measure (Analytics)
Linkedin Insight Tag (Analytics)
Google Analytics (Analytics)
Google Font API (Font scripts)
Twitter Emoji (Font scripts)
Nginx (Reverse proxies)
PHP (Programming languages)
Microsoft 365 (Email)
MailChimp for WordPress (Marketing automation)
MailChimp (Marketing automation)
Google Tag Manager (Tag managers)
Elementor (Page builders)
Quantcast Choice (Cookie compliance)
Sectigo (SSL/TLS certificate authorities)
Website
Competitors
NVIDIA
Broader GPU portfolio for cloud and edge AI; Axelera focuses on energy-efficient edge inference with specialized in-memory computing architecture.
Qualcomm
Mobile-focused AI chips; Axelera targets industrial, robotics, and defense applications with proprietary D-IMC technology optimized for latency and power efficiency.
Intel Movidius
Vision accelerators for edge AI; Axelera's in-memory computing approach provides greater energy efficiency and lower latency for general AI inference workloads.
Graphcore
Focuses on data center AI training; Axelera specializes in edge AI inference with emphasis on privacy and energy efficiency on-device.
Why this matters: Axelera AI has raised over $450M, positioning it as one of Europe's most well-funded fabless semiconductor startups. With backing from BlackRock, Samsung, and the European Innovation Council, plus a €61.6M EU grant for the RISC-V DARE project, the company is building critical infrastructure for edge AI across defense, industrial, and robotics sectors at a time when on-device AI processing is becoming essential for privacy, latency, and cost reasons.
Best for: Industrial, defense, robotics, and retail organizations deploying AI inference on edge devices where latency, privacy, and power consumption are critical constraints.
Use cases
Real-time Video Analytics in Retail
Retail stores deploy Axelera's Metis platform in security cameras to process computer vision models locally, detecting anomalies and theft without sending video to cloud servers. This reduces latency to milliseconds, eliminates privacy concerns, and cuts bandwidth costs.
Autonomous Robotics in Manufacturing
Factory robots use Axelera chips to run real-time object detection and navigation models on-device. The low power consumption extends battery life and enables responsive autonomous behavior without cloud dependency.
Defense and Public Safety Applications
Military and law enforcement agencies deploy Axelera's platform in drones, surveillance systems, and sensor networks for secure, low-latency AI inference without transmitting sensitive data over networks.
Precision Agriculture Monitoring
Agricultural drones and ground sensors use Metis to analyze crop health, pest detection, and irrigation needs in real-time at the field edge, reducing latency and enabling immediate field decisions.
Alternatives
NVIDIA Jetson
Broader ecosystem and software maturity; Axelera offers superior energy efficiency and latency through specialized in-memory computing for specific edge AI workloads.
Google Coral
Focus on mobile and IoT devices; Axelera targets higher-performance industrial and defense applications requiring more compute capability and custom integration.
Qualcomm Snapdragon
Mobile and consumer-focused SoCs; Axelera provides specialized industrial-grade AI acceleration with emphasis on enterprise deployments and robustness.
FAQ
What does Axelera AI do? +
Axelera AI develops purpose-built hardware (Metis™ AI Platform) and software for accelerating AI inference at the edge. Using proprietary digital in-memory computing technology, their chips deliver high-performance, energy-efficient AI processing on devices rather than in the cloud, reducing latency and power consumption by 10x compared to standard computing.
How much does Axelera AI cost? +
Pricing is not publicly available. Axelera uses an enterprise/custom pricing model. Metis PCIe and M.2 cards can be ordered through their partner store, but specific pricing requires direct contact with the company or authorized distributors.
What are alternatives to Axelera AI? +
Key alternatives include NVIDIA Jetson (broader ecosystem, broader use cases), Google Coral (mobile/IoT focus), Qualcomm Snapdragon (consumer devices), and Intel Movidius (vision-specific). Each has different strengths; Axelera excels in energy efficiency and latency for industrial and defense edge AI applications.
Who uses Axelera AI? +
Axelera serves industrial manufacturers, defense and public safety agencies, robotics companies, retail enterprises, and agritech firms. The company has shipped to 500+ global customers across these sectors, though specific named customers are not publicly disclosed.
How does Axelera AI compare to NVIDIA? +
NVIDIA Jetson offers a broader software ecosystem and maturity across cloud-to-edge AI workloads. Axelera specializes in energy-efficient, low-latency edge inference using proprietary in-memory computing architecture, making it ideal for battery-powered devices and latency-critical industrial applications where NVIDIA's broader approach may consume more power.
Tags
edge AI
AI inference
semiconductor
hardware acceleration
in-memory computing
RISC-V
industrial AI
robotics AI
energy efficiency
real-time processing