Unleashing Edge AI Potential: Eta Compute's New Collaboration with NXP Semiconductors
November 29, 2023 -- We are pleased to announce that Eta Compute and NXP® Semiconductors have partnered together to enable the rapid development of Edge AI products using Aptos – Eta Compute’s leading MLOps platform. Edge AI represents the convergence of two vastly different worlds – AI and embedded systems – creating new enablement challenges. A key to successfully bringing these worlds together is collaboration and integration. NXP Semiconductors is a leader in secure and connected solutions for embedded applications, and Eta Compute’s expertise is in optimizing AI applications for the deployment at the edge.
At the core of this collaboration is the onboarding of NXP’s advanced AI-focused chips and software tools onto the Aptos platform, a highly productive, no-code MLOps toolchain that simplifies model development, deployment, and management for low power edge processors. Eta Compute has completed the integration of NXP’s new MCX N Series microcontroller, which includes their eIQ® Neutron Neural Processing Unit (NPU) – a highly scalable, area and power efficient machine learning accelerator core architecture. This integration is not merely a handoff between NXP’s eIQ machine learning software enablement and the Aptos platform; it’s a powerful synergy that allows Aptos to deeply understand the AI capabilities of the NXP solution. In doing so, Aptos can optimize and adapt AI models specifically for this chip, resulting in an unparalleled level of model efficiency and performance that was previously elusive and time consuming.
This collaboration marks a significant step towards closing the gap between AI and embedded systems. By eliminating the need to intricately understand the specifics of a chip’s capabilities and constraints, Aptos empowers both ML developers and embedded systems engineers. AI developers can now create and deploy optimized ML models with ease, leveraging the power of NXP’s chips without deep knowledge of embedded techniques to manage tight resources such as on-chip memory and low power requirements. Embedded systems engineers can easily obtain optimal AI models using no-code toolchains. This “simplification + optimization” is at the heart of making the Edge AI revolution real — making it more accessible and efficient than ever before.
In the fast-evolving world of Edge AI, collaboration is key to success. Eta Compute and NXP Semiconductors’ partnership signifies the industry’s commitment to pushing the boundaries of what’s possible in Edge AI by breaking down barriers and providing effective solutions for product developers. With this collaboration, the future of Edge AI looks more promising than ever. Stay tuned for further breakthroughs as we continue to bridge the gap between AI and embedded systems.
You can join our Aptos beta program for free, here.
Related Semiconductor IP
- Temperature Glitch Detector
- Clock Attack Monitor
- SoC Security Platform / Hardware Root of Trust
- SPI to AHB-Lite Bridge
- Octal SPI Master/Slave Controller
Related News
- Eta Compute Debuts with World's Lowest Power Microcontroller IP Targeting Energy Harvesting Segment
- Eta Compute Announces New Reference Design for EtaCore ARM Cortex-M3, the World's Lowest Power Microcontroller IP
- Eta Compute's David Baker, Ph.D. To Present On Low-Power IP For The Internet of Things At Arm TechCon October 25
- Eta Compute Secures $8 Million Series A Financing to Accelerate Development of the 3rd Generation of Neural Networks
Latest News
- RaiderChip NPU for LLM at the Edge supports DeepSeek-R1 reasoning models
- The world’s first open source security chip hits production with Google
- ZeroPoint Technologies Unveils Groundbreaking Compression Solution to Increase Foundational Model Addressable Memory by 50%
- Breker RISC-V SystemVIP Deployed across 15 Commercial RISC-V Projects for Advanced Core and SoC Verification
- AheadComputing Raises $21.5M Seed Round and Introduces Breakthrough Microprocessor Architecture Designed for Next Era of General-Purpose Computing