Synaptics Launches Torq Edge AI Platform Featuring Google’s RISC-V Coral Open NPU, Boosting Multimodal Inference for Embedded IoT

September 6, 2026

Edge AI and Vision Insights USA

Overview

Synaptics unveiled its “Torq™ Edge AI platform,” a high-performance, low-power Neural Processing Unit (NPU) subsystem for edge AI, integrated with Google’s RISC-V based Coral Open NPU. Part of the Astra Machina SL2610 Development Kit, this platform efficiently executes modern AI models, including vision transformers and multimodal pipelines, without proprietary toolchain constraints. It directly brings multimodal, transformer-class inference to embedded IoT hardware, significantly enhancing edge AI and visual intelligence.

In Depth

Key Findings

Synaptics has introduced the “Torq™ Edge AI platform,” a high-performance, low-power Neural Processing Unit (NPU) subsystem for edge AI, integrating Google’s RISC-V based Coral Open NPU. This new platform is designed to dramatically enhance edge AI and visual intelligence by bringing multimodal and transformer-class AI inference capabilities directly to embedded IoT hardware.

Technical Details

The “Torq™ Edge AI platform” is a core technology within Synaptics’ Astra Machina SL2610 Development Kit. A key feature is its open design philosophy, which allows for the efficient execution of a wide range of modern AI models without being restricted by proprietary toolchains or specific model constraints. This flexibility enables seamless integration of today’s complex AI applications, such as vision transformers and multimodal pipelines that process multiple data formats. The integrated Google RISC-V based Coral Open NPU provides exceptional power efficiency and processing speed, facilitating advanced AI inference in resource-constrained edge environments. This empowers embedded IoT devices like smart cameras, industrial robots, smart home appliances, and wearables with more sophisticated object recognition, facial authentication, gesture recognition, voice command processing, and anomaly detection capabilities.

Background & Context

Edge AI offers significant advantages by processing AI directly on devices without transferring data to the cloud, including reduced latency, enhanced privacy protection, and bandwidth savings. With the proliferation of IoT devices, there’s a rapidly increasing demand for real-time and intelligent processing at the edge. However, traditional embedded systems have struggled to run complex AI models, especially computationally intensive transformer and multimodal models. Synaptics’ Torq Edge AI platform addresses this challenge by combining the openness of the RISC-V architecture with Google Coral’s AI acceleration technology, aiming to elevate edge device AI capabilities to levels comparable with cloud AI.

Strategic Significance & Outlook

This new platform represents a crucial milestone in accelerating innovation within the edge AI market. Developers will be able to more easily integrate powerful and flexible AI engines into embedded products, leading to a dramatic improvement in smart device functionality. Furthermore, the adoption of an open NPU architecture is expected to foster the growth of the AI ecosystem and attract contributions from a diverse developer community. Synaptics is poised to lead the next generation of intelligent edge computing solutions across a wide range of sectors, including industrial, consumer, and automotive.

Source: https://www.edge-ai-vision.com/2026/09/edge-ai-and-vision-insights-september-2-2026/

Get our weekly technology intelligence — free

Receive an infographic that lets you judge at a glance whether each field’s analysis report is worth reading.

Subscribe Free — Weekly Tech Intelligence

By subscribing, you’ll receive Troy-Technical’s weekly technology intelligence newsletter.