San Jose, CA, October 20, 2026 – Synaptics® Incorporated (Nasdaq: SYNA) today announced that the latest version of the Synaptics Coralboard is available for purchase through DigiKey starting today. Developed in collaboration with Google Research, the Coralboard SL is powered by the Synaptics Astra® SL2619, which integrates the 1 TOPS Synaptics Torq™ NPU featuring the first implementation of the Coral NPU. Following an initial limited-edition run, this version moves the platform to general availability through DigiKey’s global distribution network. The Coralboard portfolio is built for ultra-low power, always-on applications using ambient sensing, and is designed to deliver efficient, accelerated on-device inference for resource-constrained form factors.

Expanding the ecosystem and broadening developer choice, Synaptics also introduced the Coralboard SR, based on the newly launched Synaptics Astra® SR200 Series of xMCUs, a new class of AI-native, multimodal processors combining heterogeneous compute with integrated high-bandwidth memory to bring vision, voice, and graphics to Edge and Physical AI systems. 

The Coralboard family is designed to deliver a consistent Edge AI developer experience across the SL and SR versions thanks to the MLIR-based Synaptics Torq™ compiler and runtime. Synaptics’ Torq open-source toolchain supports widely used ML frameworks, giving developers a single path from model optimization through deployment. With an open, collaborative development and continuous refinement approach, Torq is designed to deliver day-0 support for Edge AI model families, such as Google Gemma3 270M (achieves up to 12-16 tokens/second with INT8 and 4-bit quantization respectively), LiquidAI LFM2.5, and Ultralytics YOLO26, employing active-research techniques such as batch-prefill and dynamic quantization.

Torq enables optimal execution across Synaptics Astra processor portfolio by natively mapping model network layers to accelerator engines. Additionally, the open-source nature of Torq allows users to deploy agentic AI tools and workflows that parse through the source code and assist in feature-addition, performance customization, and debugging, promoting true Edge AI developer freedom. 

The Synaptics Coralboard brings together the open hardware and software foundation for private, efficient Edge AI applications across wearables, hearables, smart home appliances, automation hubs, industrial control systems, and robotics.

Synaptics Coralboard SL comes with: 

• Astra SL2619 with 2x 2GHz Arm® Cortex® A55, Cortex-M52, 1 TOPS Torq/Coral NPU  
• 2GB DDR4, MicroSD card slot, NOR flash memory
• Camera and display support via CSI/DSI and USB
• QWICC and mikroBUS connectors
• Add-on HATs for HDMI display, power, camera, and digital mics
• Wi-Fi®/Bluetooth® connectivity through an M.2 expansion slot 

Synaptics Coralboard SR offers: 

• Astra SR250 Series with 800MHz Arm Cortex M85, Cortex M52, 400 GOPS Torq/Coral NPU  
• Integrated SRAM up to 8MB and high-bandwidth custom DRAM up to 256MB
• MicroSD card slot, NOR flash memory
• Multi-camera and display support via CSI/DSI and USB, dual ISPs, digital mics
• QWICC and mikroBUS connectors
• Wi-Fi/Bluetooth connectivity

Building on its work with Google Research, Synaptics extends the Coralboard portfolio with a shared goal to defragment the developer experience for Edge AI. 

“Our collaboration with Synaptics started from a shared view of what open Edge AI hardware should look like, and the Coralboard SL was the first developer-ready solution we co-created,” said Billy Rutledge, Director, Google Research. “The Synaptics Astra SR200 series now brings the second generation of the Coral NPU IP with RISC-V vector core support into silicon, and the Coralboard SR puts that in developers’ hands. Each version widens what can run at the Edge without asking developers to start over.”

"The Synaptics Coralboard is a starting line the whole ecosystem can share,” said Olivier Lacombe, Product Director, Google DeepMind. “With Gemma models optimized and ready to run on Coralboard, developers can stop solving the same integration puzzle, scale the model as their application requirements evolve, and start building the things only they can imagine. 

“The Synaptics Coralboard was designed to address a developer need for an open, cost-optimized AI-native Edge platform for the IoT,” said Vikram Gupta, SVP & GM, Edge Compute and Connectivity Solutions, Synaptics. “The limited-edition run validated this thinking, and our partnership with DigiKey now puts the Coralboard within reach of engineers worldwide. Integrating the open-source Synaptics Torq™ compiler, AI/ML developers get a unified, consistent experience that scales across our Astra portfolio.” 

“Edge AI is where the next wave of design activity is happening, and developers need a starting point they can trust,” said Mike Slater, Vice President of Global Business Development, DigiKey. “As the exclusive distribution partner for the Synaptics Coralboard, we are putting an open, production-ready design platform in front of millions of engineers in our global network.” 

“We’re excited to see Liquid.AI’s LFM2.5 Edge AI models running on the Synaptics Astra® SL2619. Our models are going to help drive new opportunities for Edge AI,” said Alan Lau, Head of Silicon Partnerships at Liquid AI.

“NPU compilers have largely been closed, black-box toolchains, which makes an open-source compiler like Synaptics Torq™ stand out in our industry,” said Daniel Chang, CEO, ENERZAi. “As a contributor to the Torq project, we've seen first-hand how its open, MLIR-based design gives developers the freedom to inspect, extend, and optimize models to accelerators. At the same time, as an Edge AI software company commercializing 1.58-bit on-device AI agents, we see Torq as a flexible and compelling path to bring these models into real-world products.” 
 

Availability

The Synaptics Coralboard is available to order from DigiKey starting October 20, 2026. For product details, documentation, and to get started, visit www.synaptics.com/coral-dev-board.