Specifications
Product
Synaptics Astra Machina SL2610 Development Kit
Family
Astra Machina Foundation Series
SoC
Synaptics Astra SL2619 (SL2610 product line)
Product Line
SL2611, SL2613, SL2615, SL2617, SL2619 — pin-to-pin compatible
CPU
Single / dual-core Arm Cortex-A55 (up to 2 GHz)
System Manager
Arm Cortex-M52 with Helium (up to 200 MHz)
NPU
1 TOPS transformer-capable NPU (Torq T1)
ML Core
Google Coral NPU — RISC-V based, open-source
AI Platform
Synaptics Torq Edge AI — IREE / MLIR compiler and runtime
GPU
Arm Mali-G31 3D GPU (OpenGL / Vulkan)
Camera
MIPI-CSI, 2160p30 with HDR; MIPI-DSI display
Audio
3 × TDM/I2S, 16 channels, support for 8 digital microphones
Connectivity
Synaptics Veros Wi-Fi 6/6E/7, BT/BLE, Thread, UWB options
Networking
Dual Gigabit Ethernet with IEEE 1588 and TSN
Industrial I/O
2 × CAN-FD ports
Kit Architecture
Core compute module + I/O base board + daughter cards (Wi-Fi/BT, debug, programmable I/O)
OS Support
Yocto Linux, Android
RTOS Support
FreeRTOS, Zephyr on the Cortex-M52 system manager
Target Applications
Multimodal AI-native IoT, industrial vision, battery-powered edge devices
Availability
SL2610 general availability planned Q2 2026
Lead Time
4–6 weeks
MOQ
1 unit
Overview
The Synaptics Astra Machina SL2610 development kit is the evaluation path for the SL2610 product line — five pin-to-pin compatible SoC families (SL2611, SL2613, SL2615, SL2617, SL2619) spanning battery-powered passively cooled devices up to high-performance industrial vision systems.
Architecturally the kit is split into a core compute module, an I/O base board, and daughter cards for Wi-Fi/Bluetooth connectivity, debug and programmable I/O. That modularity is deliberate: it lets a design start on the full dev kit and then carry the same compute module into production hardware.
The AI story is the open one. The 1 TOPS transformer-capable Torq T1 NPU sits alongside the first production implementation of Google's open-source RISC-V Coral NPU, driven by an IREE/MLIR compiler and runtime rather than a proprietary toolchain — which matters to teams who want to avoid vendor lock-in on the edge AI stack.
For industrial use the SL2610 adds dual Gigabit Ethernet with IEEE 1588 and TSN plus two CAN-FD ports, so it interfaces directly to deterministic production networks without a bridge board.
Key Benefits
Open AI stack — Google Coral NPU (RISC-V) with IREE/MLIR, no proprietary compiler lock-in; modular kit — compute module + I/O base + daughter cards carries through to production; industrial networking built in — dual GbE with TSN and 2 × CAN-FD; 5 pin-compatible SoC families for scaling a single design.
Applications
Multimodal AI-native IoT product development, industrial vision prototyping, battery-powered edge inference devices, robotics and building-automation gateways, and any design needing deterministic TSN Ethernet plus CAN-FD at the edge.
Request a Quote — SYNAPTICS ASTRA MACHINA SL2610 DEV KIT — CORAL NPU EDGE AI PLATFORM
QS Compute — global B2B supply of AI computing hardware, edge AI systems and accelerators. Volume pricing, 15-day sample lead time.
Get Your Quote →