-

Efinix® Launches Titanium Edge™ FPGA Family, Solving the Demanding Requirements of Edge AI

New family delivers breakthrough power reduction, System-in-Package integration, high-speed MIPI I/O, advanced SEU scrubbing, and post-quantum security for constrained edge AI deployments

CUPERTINO, Calif.--(BUSINESS WIRE)--Efinix® Inc., the FPGA pioneer accelerating edge AI innovation, today announced the Titanium Edge™ family of FPGAs, purpose-built for the most demanding edge AI applications. Built on the company’s proven Titanium architecture, Efinix’s latest suite of devices is optimized for next-generation edge AI deployments across several high-growth markets.

“In edge AI applications, every milliwatt, every square millimeter, and every potential attack surface matters,” said Sammy Cheung, co-founder and CEO of Efinix. “Titanium Edge reflects years of co-development with customers, pushing the limits of what is achievable in constrained deployments. We are giving designers the performance they need without forcing them to compromise on power, footprint, or security.”

With a tightly integrated set of silicon innovations that raise the bar for what an FPGA can deliver in real-time AI deployments, Titanium Edge addresses five key challenges facing AI system designers:

  • Power – Titanium Edge delivers 50% lower static power than the Titanium family, which already set a leading benchmark for FPGA power efficiency. This enables always-on AI inference and processing in highly power- and thermally-constrained edge systems.
  • Space – System-in-Package (SiP) devices integrate FPGA, HyperRAM, and boot flash die into a single module, cutting PCB footprint by up to 60% versus discrete designs. Packages scale down to 5.5 mm × 5.5 mm, enabling ultra-compact edge deployments.
  • Performance – Titanium Edge supports up to 2.5 Gbps I/O with MIPI interfaces in x1/x2/x4/x8 data lane widths, enabling multi-sensor AI pipelines for vision, imaging, and autonomous systems without requiring external bridge components.
  • Reliability – Titanium Edge is designed for high-reliability edge deployments where system uptime and data integrity are critical. A dedicated hardware SEU scrubbing engine continuously monitors and corrects configuration memory to protect against radiation-induced bit flips.
  • Security – Security variants add a post-quantum-ready cryptography aligned with CNSA 2.0 and EU Cyber Resilience Act timelines. Key features include ML-DSA-65 (CRYSTALS-Dilithium) signature authentication, AES-256-GCM encryption, RSA-4096 / EC-DSA-384 support, PUF-based key derivation, secure boot and chain-of-trust, TRNG for entropy generation, full key lifecycle management (including rotation, revocation, and anti-rollback), and secure debug and JTAG disable support.

“Titanium Edge is the culmination of close collaboration with customers building next-generation edge AI systems in sectors where failure is not an option,” said Tony Ngai, co-founder, Co-President, and CTO of Efinix. “The integration of post-quantum security directly into silicon, combined with hardware SEU scrubbing and SiP packaging, gives our customers a platform they can design once and trust for the full product lifecycle—from certification through decade-long field deployment.”

Titanium Edge Device Family

The Titanium Edge family spans multiple logic densities and package configurations, including ultra-compact devices for space-constrained designs and higher-capacity variants for sensor fusion and AI inference workloads.

  • Ti125 — 123,000 Logic Elements (Sampling Now)
  • Ti125 SiP — 123,000 Logic Elements with integrated 512Mb HyperRAM and SPI boot flash (Sampling August 2026)
  • Ti95 — 93,000 Logic Elements (Sampling Now)
  • Ti70 — 68,000 Logic Elements (Sampling Q4 2026)
  • Ti70 SiP — 68,000 Logic Elements with integrated 256Mb HyperRAM and SPI boot flash (Sampling Q4 2026)
  • Ti40 — 39,000 Logic Elements (Sampling Q4 2026)

Security variants of the Ti125 and Ti95 are planned for Q4 2026 sampling.

All Titanium Edge devices are supported by the Efinity® Integrated Development Environment, delivering a complete RTL-to-bitstream design flow. The broader Efinix ecosystem also includes Sapphire SoC solutions for embedded RISC-V AI processing, along with FPGA fabric-based post-quantum cryptography accelerators that support ML-DSA, ML-KEM, SHA-3, and SHAKE algorithms.

For evaluation board availability, design support, and pricing, contact your local Efinix sales representative or visit www.efinixinc.com.

About Efinix

Efinix® accelerates edge AI innovation through its disruptive FPGA architecture and solutions. The company delivers energy-efficient, high-performance FPGAs for AI-driven markets including industrial automation, machine vision, automotive systems, physical AI, and edge computing. Efinix provides a complete development ecosystem, including the Efinity® Integrated Development Environment for FPGA design from RTL to bitstream, and Sapphire SoC solutions for embedded processor and AI acceleration from C/C++ and AI model garden to execution. For more information, visit www.efinixinc.com.

Contacts

Media Contact:
Tyler Weiland
Shelton Group
+1-972-571-7834
tweiland@sheltongroup.com

Company Contact:
Robert Beachler
Efinix Inc.
bobb@efinixinc.com

Efinix


Release Versions

Contacts

Media Contact:
Tyler Weiland
Shelton Group
+1-972-571-7834
tweiland@sheltongroup.com

Company Contact:
Robert Beachler
Efinix Inc.
bobb@efinixinc.com

Back to Newsroom