-- In an era where AI models evolve faster than hardware lifecycles, industrial device manufacturers face a growing dilemma - the AI treadmill trap. Every six to twelve months, new neural architectures and large models emerge, leaving once-cutting-edge designs behind. Yet industrial systems, from factory HMIs to medical diagnostics interfaces, must remain certified and stable for a decade or more.
SolidRun, in collaboration with Renesas, is tackling this challenge head-on with a modular platform approach built around the Renesas RZ/G2L microprocessor - delivering stable HMI, isolated from upgradable intelligence.
Breaking Free from the AI Treadmill
AI processing requirements continue to evolve at breakneck speeds. Transformer-based vision models, generative diagnostics, and adaptive energy optimization now demand compute resources that were unthinkable just a few years ago. Traditional monolithic designs force manufacturers into costly full-system redesigns and possibly re-certifications whenever AI algorithms change.
By contrast, SolidRun’s design philosophy - anchored by the RZ/G2L System-on-Module (SoM) - decouples the stable control and display layer from rapidly advancing AI compute. The result:
- HMI and control systems stay stable for 5-10+ years
- AI module can be upgraded as new accelerators, NPUs, or companion SoMs emerge
This architecture turns what some view as a hardware limitation - the RZ/G2L’s lack of an integrated AI accelerator - into a strategic advantage for long-lifecycle industrial systems.
A Stable Foundation: The Renesas RZ/G2L
Built on Arm Cortex-A55 and Cortex-M33 cores, the RZ/G2L provides the ideal long-term foundation for industrial, medical, and building-automation HMIs. It combines:
- High-efficiency processing with low thermal footprint
- Secure boot, TrustZone, and long-term Linux/Yocto support
- Multimedia, 3D graphics, and camera interfaces for advanced UX
Because of its stable silicon roadmap and long lifecycle support, the RZ/G2L can serve as a permanent platform layer - while AI capabilities evolve independently above it.
SolidRun’s Modular Architecture: Built for Continuous Innovation
The SolidRun RZ/G2L SoM integrates high-speed DDR4, industrial-grade components, and flexible connectivity into a compact, production-ready module. It’s engineered for modularity, enabling manufacturers to:
- Deploy stable HMIs today, without locking into fixed AI performance
- Add or replace companion modules to introduce new AI inference engines or co-processors
- Maintain certification and safety compliance without full product redesign
From robotic interfaces to energy dashboards and medical imaging terminals, this separation of roles allows continuous improvement without compromising reliability.
Engineering for the Edge – and for Longevity
SolidRun backs its hardware with deep engineering support, BSPs, and reference carrier boards that ensure long-term software consistency. This stability is key in environments where downtime costs are measured in thousands per minute and regulatory recertification is
non-trivial.
The result: industrial AI that evolves on its own terms, not the market’s schedule.
A Shared Vision for Intelligent Infrastructure
Together, Renesas and SolidRun are redefining what embedded intelligence means for the industrial edge - not by chasing every new AI accelerator, but by designing systems that can adapt gracefully.
Whether upgrading a factory HMI with new vision models, refreshing hospital diagnostics with next-gen algorithms, or optimizing building automation, the RZ/G2L SoM makes future AI innovation a matter of modular evolution, not costly reinvention.
About SolidRun
SolidRun is a global provider of embedded and edge computing solutions known for its modular design philosophy and open-source commitment. By partnering with Renesas and other industry leaders, SolidRun empowers OEMs to build the next generation of intelligent, reliable, and upgradable industrial systems.
Contact Info:
Name: Ata Ziv
Email: Send Email
Organization: SolidRun
Website: https://www.solid-run.com/
Release ID: 89175958
In case of identifying any problems, concerns, or inaccuracies in the content shared in this press release, or if a press release needs to be taken down, we urge you to notify us immediately by contacting error@releasecontact.com (it is important to note that this email is the authorized channel for such matters, sending multiple emails to multiple addresses does not necessarily help expedite your request). Our dedicated team will be readily accessible to address your concerns and take swift action within 8 hours to rectify any issues identified or assist with the removal process. We are committed to delivering high-quality content and ensuring accuracy for our valued readers.
