Up to 5x
performance per dollar
Cost efficiency with no performance compromise
Edge Efficiency,
Data Center
Performance.
Vision and Gen AI running on the world's most efficient silicon, from embedded to edge to server
Axelera's technology
Digital in‑memory computing.
Compute inside SRAM. The data stops travelling, the watts stop burning.
15
TOPS per watt
Demanding workloads without the power, the cost or the heat.
99.9%
relative accuracy
Post-training quantization preserves FP32-equivalent accuracy with no retraining
Axelera products
One architecture, three ways to deploy.
Digital in-memory computing
The array becomes the chip
Inside our cores
Multi-core architecture, with all the compute you need
No compromise
99.9% accuracy on the most demanding AI workloads
Scaling
Chips that work together
A card for every use-case
Embedded to edge to server
Software Development Kit
One platform for any model
Scale
Deploy from robot to rack
From silicon to data center
One chip becomes the whole stack.
Chip, card, server, rack, data center: one architecture at every scale.
Three ways to deploy.
Embedded
AI in the machine: M.2 modules and single-board computers
Explore Embedded →
Edge
AI next to your data: PCIe cards for edge servers and industrial PCs
Explore Edge →
Server
AI at data center scale: multi-AIPU cards that bring inference density to racks and supercomputers.
Explore Server →Examples
Axelera in action
Real-use cases running on Axelera silicon
TILED INFERENCE high-resolution scenes sliced into sections, so small objects stay detectable
SEGMENTATION AND DETECTION self-checkout: shopper, basket, bags and products masked and labelled, 18.6 to 22.6 fps end to end
PPE DETECTION hard-hat compliance across a wall of 24 live streams, with the host CPU at 4%
LLM, illustration an on-device language model answering across twelve camera streams
Silicon
Three generations, one architecture
One architecture that spans embedded to rack-scale deployments
Shipping
Metis®
Quad-core AIPU
214 TOPS per chip
- Average power
- 4-8 W
- Memory
- up to 16 GB per chip
- LLM capacity
- up to 8B parameters
- YOLOv8s
- 653 FPS, see note
- ResNet-50
- up to 3,200 FPS
Next generation
Europa™
Double the cores, sixteen vector units, pre/post-processing on die.
629 TOPS per chip
- Average power
- 45 W
- Memory
- up to 64 GB per chip
- LLM capacity
- up to 32B parameters
- YOLOv8s
- 2,850 FPS · 4.36× Metis
- Architecture
- 8× AIPU cores
Announced
Titania™
A chiplet that scales the architecture into HPC.
Same job. Different silicon.
Benchmark figures
Throughput
Ultralytics YOLOv7 at 640 x 640, frames per second, higher is better
Competitor
0
Axelera Metis
0
2.1x the competitor throughput Axelera publishes
Across the six networks with a published competitor figure: SSD-MobileNet v2 2,024 against 784; YOLOv5m 456 against 156; YOLOv7 212 against 100; YOLOv8s 610 against 491; YOLOv8m 235 against 149; YOLOv8l 181 against 64.
Cost efficiency
ResNet50v1 frames per second per dollar, higher is better
Competitor
0
Axelera Metis
0
3x the frames per second per dollar shown
Energy efficiency
TOPS per watt at INT8, higher is better
Competitor
0
Axelera Metis
0
4x the TOPS per watt shown
Throughput is measured host-side on a system with an Intel Core i9-13900K, and Axelera notes that the competitor data was obtained from public sources as of August 2026. The three-generation table above quotes Axelera’s accelerator comparison, which lists the same network at 653 FPS. Metis AIPU benchmarks
Embedded · Edge · Server
From Chip to Complete System
Choose how you deploy.
Next generation
Europa™ PCIe card (announced)
PCIe card · next-generation AIPU · not yet shipping
Learn more →Solutions
One platform, multiple end-users
By industry
Industrial Manufacturing
Edge AI for machine vision and Industry 4.0 equipment, raising productivity in production lines and warehouses.
Learn more →
Retail
Real-time analytics across every customer touchpoint, so inventory control is easier and shrinkage reduces.
Learn more →
Security
Video analytics that turns a camera from a video source into a smart sensor for investigation and risk prevention.
Learn more →
Healthcare
High-performance AI at the edge keeps patient data safer while filling critical gaps in healthcare systems.
Learn more →
Smart Cities
Technology and data-driven solutions that make cities more efficient, sustainable and responsive to residents.
Learn more →
Agriculture
Better crop yields, optimized nutrients in the soil, and the right amount of water reaching the roots.
Learn more →
Space
Autonomous rovers navigating distant planets, and AI-powered satellites monitoring Earth and deep space.
Learn more →By application
Computer Vision
Vision moved to where the data is created: real-time processing, less data movement, scale into remote sites.
Learn more →
Robotics
Robots that enter non-permissive environments and make real-time decisions with no internet connection.
Learn more →
High Performance Compute
Digital In-Memory Compute scales up: the Titania chiplet serves the high performance compute market directly.
Learn more →From model to production, powered by the Voyager® SDK
Everything you need to build, faster.
One toolchain, every deployment: describe your pipeline in YAML and the Voyager SDK compiles, quantizes and runs it on any Axelera product line, with no retraining required.
Voyager SDK: the end-to-end software stack
Purpose-built for AI inference: a declarative YAML pipeline language, 100+ supported models through one toolchain, and API compatibility with industry standards. From idea to deployment quickly, with deep customization when you need it.
Voyager Wingman: the agentic AI platform
Purpose-built for Axelera hardware with deep, curated knowledge of the SDK and documentation: Voyager Wingman builds, debugs and optimizes inference pipelines from natural language, at 20 to 30% higher accuracy than general-purpose AI coding tools.
Partner Accelerator Network
Complete solutions, together.
From hardware to software to distribution, Axelera partners with leading companies to deliver complete AI solutions.
What's new
Latest news.
Partnership
ASRock Industrial partners with Axelera on AI inference
Metis AIPU cards integrate into ASRock Industrial edge platforms for compact, scalable inference.
Read the storyProduct line
Same amazing products, updated naming
The line-up aligns around Embedded, Edge and Server designations: same technology, clearer names.
Read the storyLaunch
Introducing the Axelera AI Mini PC
Production-grade AI vision inference in a compact form factor, powered by the Embedded 113m.
Read the storyDevelopers first
Built with the community.
Developers, integrators and researchers building on Axelera hardware swap pipelines, benchmarks and answers every day. The messages shown here are illustrations of that traffic.
Join the communityWhat do you need?
Not sure which card?
Find the right card with Voyager Wingman.
Describe the workload, the box and the power constraints. Voyager Wingman knows the line-up.
Open the chat →Ready to test
Go straight to the shop.
Embedded M.2 modules, compute boards, and Edge and Server PCIe cards, ready to order today.
Order hardware →Rather talk to a human? Contact sales.
Talk to us