• Home
  • About
  • Events Calendar
  • Blueprint Guidelines
  • Privacy Policy
  • Manage Email Delivery
  • NextGenInfra.io
No Result
View All Result
Converge Digest
Wednesday, June 10, 2026
  • Home
  • About
  • Events Calendar
  • Blueprint Guidelines
  • Privacy Policy
  • Manage Email Delivery
  • NextGenInfra.io
No Result
View All Result
Converge Digest
No Result
View All Result

Home » Axelera Launches Europa AIPU with 629 TOPs 

Axelera Launches Europa AIPU with 629 TOPs 

October 24, 2025
in All
A A

Axelera AI introduced its new Europa AI Processing Unit (AIPU), a high-performance, power-efficient, and cost-optimized accelerator designed for multi-modal AI workloads spanning edge to enterprise environments. Featuring RISC-V vector processors and Axelera’s proprietary Digital In-Memory Compute (D-IMC) technology, Europa delivers up to 629 TOPs at INT8 precision within a compact 35×35 mm package. Shipments of both the AIPU and the PCIe card variants are scheduled for the first half of 2026.

The Europa AIPU architecture combines eight second-generation AI cores with 16 RISC-V vector processors to handle AI and non-AI pre- and post-processing tasks. With 128 MB of on-chip L2 SRAM and a 256-bit LPDDR5 interface delivering 200 GB/s of bandwidth, the chip eliminates traditional memory bottlenecks while maintaining up to 5× greater performance-per-watt and performance-per-dollar than comparable solutions. Europa integrates an H.264/H.265 decoder for simultaneous visual, language, sensor, and speech processing, allowing the host CPU to focus on higher-level application logic.

The Europa PCIe card version supports configurations from 16 GB (single-chip) to 256 GB (quad-chip), enabling scalable local AI processing and real-time decoding of multiple 4K video streams. Both models leverage the company’s Voyager SDK for seamless development and deployment across the Europa and Metis product families. Europa also includes a secure enclave for enterprise-grade workload protection.

  • 629 TOPs INT8 performance using eight second-generation AI cores
  • 16 RISC-V vector processors delivering 4915 GOPS for non-AI workloads
  • 128 MB on-chip SRAM and 200 GB/s memory bandwidth
  • Integrated H.264/H.265 decoder and PCIe Gen4 interface
  • Multiple form factors, including compact 35×35 mm package and scalable PCIe cards
  • Voyager SDK compatibility across Europa and Metis accelerators

“Organizations shouldn’t have to choose between the raw compute power they want and the real-world usability they need,” said Fabrizio del Maffeo, CEO of Axelera AI. “Europa makes enterprise-class AI processing power available to nearly anyone.”

🌐 Analysis:

Axelera’s Europa AIPU positions the company as a strong European challenger in the edge-to-data-center AI accelerator market. By combining in-memory compute with open RISC-V vector processing and high integration density, Europa directly targets segments dominated by NVIDIA Jetson, Intel Movidius, and AMD’s embedded AI platforms. Its focus on sovereignty and local manufacturing aligns with EU initiatives like DARE and the European Chips Act, suggesting Axelera could become a foundational player in Europe’s AI hardware ecosystem.

🌐 We’re tracking the latest developments in AI infrastructure and semiconductor acceleration. Follow our ongoing coverage at: https://convergedigest.com/category/semiconductors/

🌐 We’re launching the “Data Center Networking for AI” series on NextGenInfra.io and inviting companies building real solutions—silicon, optics, fabrics, switches, software, orchestration—to share their views on video and in our expert report. To get involved, send a note to [email protected] or [email protected].

ShareTweetShareSummarizeSummarize
Previous Post

PECC Summit: NVIDIA’s Ashkan Seyedi on AI Networking

Next Post

Verizon: AI-Powered Mobile Threats Escalate

Jim Carroll

Jim Carroll

Editor and Publisher, Converge! Network Digest, Optical Networks Daily - Covering the full stack of network convergence from Silicon Valley

Related Posts

Optical

Colt and Ciena Achieve 800GbE Quantum-Safe Across the Atlantic

June 10, 2026
All

AWS Launches Graviton5 CPU with 192 Cores for Agentic AI

June 10, 2026
Research

Dell’Oro: AI Infrastructure Spending Pushes 2026 Data Center Capex Above $1 Trillion

June 10, 2026
Semiconductors

TDK Acquires Fabric8Labs to Scale Advanced Cooling for AI Data Centers

June 10, 2026
Quantum

Xanadu Sets New Benchmark for Ultra-Low-Loss Photonic Chip Packaging

June 10, 2026
Research

Dell’Oro: Campus Ethernet Switch Revenue Climbs in 1Q 2026

June 10, 2026
Next Post

Verizon: AI-Powered Mobile Threats Escalate

Categories

  • 5G / 6G / Wi-Fi
  • AI Infrastructure
  • All
  • Automotive Networking
  • Blueprints
  • Clouds and Carriers
  • Data Centers
  • Enterprise
  • Explainer
  • Feature
  • Financials
  • Last Mile / Middle Mile
  • Legal / Regulatory
  • Optical
  • Quantum
  • Research
  • Security
  • Semiconductors
  • Space
  • Start-ups
  • Subsea
  • Sustainability
  • Video
  • Webinars

Archives

Tags

5G All AT&T Australia AWS Blueprint columns BroadbandWireless Broadcom China Ciena Cisco Data Centers Dell'Oro Ericsson FCC Financial Financials Huawei Infinera Intel Japan Juniper Last Mile Last Mille LTE Mergers and Acquisitions Mobile NFV Nokia Optical Packet Systems PacketVoice People Regulatory Satellite SDN Service Providers Silicon Silicon Valley StandardsWatch Storage TTP UK Verizon Wi-Fi
Converge Digest

A private dossier for networking and telecoms

Follow Us

  • Home
  • About
  • Events Calendar
  • Blueprint Guidelines
  • Privacy Policy
  • Manage Email Delivery
  • NextGenInfra.io

© 2026 Converge Digest - A private dossier for networking and telecoms.

No Result
View All Result
  • Home
  • About
  • Events Calendar
  • Blueprint Guidelines
  • Privacy Policy
  • Manage Email Delivery
  • NextGenInfra.io

© 2026 Converge Digest - A private dossier for networking and telecoms.

This website uses cookies. By continuing to use this website you are giving consent to cookies being used. Visit our Privacy and Cookie Policy.
Go to mobile version