The LLM630 Compute Kit is an AI inference platform featuring an AX630C SoC (3.2 TOPs NPU), 4GB LPDDR4, 32GB eMMC, Wi-Fi 6, Gigabit Ethernet, dual USB-C ports, and an SMA antenna interface. It supports CV, speech, and LLM tasks with the StackFlow framework. M5Stack LLM630 Compute Kitis anAI large language model inference development platformdesigned for edge computing and smart interactive applications. This kits mainboard is equipped with an AixinAX630CSoC processor, integrating a high-efficiencyNPUwith 3.2 TOPs at INT8 computing power, providing powerful AI inference capabilities to efficiently execute complexvision (CV) and large language model (LLM) tasks, meeting the needs of various intelligent application scenarios. The mainboard is also equipped with the JL2101-N040C Gigabit Ethernet chip andthe ESP32-C6 wireless communication chip, supporting Wi-Fi 6 at 2.4 G as the devicesnetwork card, providing high-speed data transmission forbridgingfunctionality between Wi-Fi and Ethernet. Whether through a wired connection for large-scale data exchange or via wireless communication for real-time interaction with a remote server or other smart devices, this platform ensures efficient data exchange. The mainboard also integrates an SMA antenna interface to further enhance the stability and transmission range of wireless signals, ensuring stable communication in complex network environments. It comes with4GB LPDDR4memory (of which 2GB is for user usage, 2GB for hardware acceleration) and32GB eMMCstorage, supporting parallel loading and inference of multiple models for efficient and smooth task handling. Working in perfect conjunction with thebaseboard, the LLM630 Compute Kits functionalities and applicability are significantly expanded. It integrates aBMI270 6-axis sensorfor precise attitude sensing and motion detection, suitable for various dynamic applications. With the built-in NS4150B Class D amplifier andmicrophone and speaker interfaces, it supports high-quality voice input and audio output, enabling full-duplex communication for an enhanced user interaction experience. The baseboard also featuresdual Grove interfacesas well as LCD/DSI and CAM/CSIMIPI interfaces, making it convenient to expand peripherals such as displays and camera modules. Additionally, the baseboard integrates anexternal antenna interfaceand a GigabitEthernet port, providing flexible network connections and enhanced wireless performance. The devicesuser buttonsenable functions such as power on/off and mode switching, enhancing usability and interactivity. The baseboardscharging chipand reservedbattery socketsupport custom battery configurations, ensuring the platform can run stably for long periods even without external power. The integratedbattery detection chipmonitors battery status in real time. TheMicroSD card slotallows for storage expansion and will later support AI model updates. The dualType-C portsnot only support high-speed data transmission but also provideOTGfunctionality, making device connections more flexible and ensuring high efficiency in data exchange and device connectivity. LLM630 Compute Kit supports theStackFlow framework, allowing developers to realize edge AI applications with just a few lines of code and quickly start various AI tasks. The platform supports a wide range of AI applications, includingcomputer vision recognition,speech recognition,text-to-speech, andwake word recognition, and supports separate invocation or automated pipeline flow for convenience. The platform also supports models such as Yolo11 DepthAnything for vision tasks, InternVL2.5-1B multi-modal large model, Qwen2.5-0.5/1.5B Llama3.2-1B large language models, and Whisper Melotts voice models, with hot update features. It will continue to support cutting-edge large models, empowering intelligent recognition and analysis, ensuring the platform keeps pace with technological developments and community trends. LLM630 Compute Kit is suitable forsecurity monitoring,smart retail,smart agriculture,smart home control,interactive robotics, andeducation, providing powerful computing capabilities and flexible expandability for edge AI applications. The StackFlow framework is an open-source AI framework developed for the Module LLM and LLM630 Compute Kit. It integrates AI algorithms such as speech, vision, and large language models. Features Aixin YuanZhi AX630C SoC 3.2TOPs @ INT8 inference performance Gigabit Ethernet communication ESP32-C6 network card Display interface (DSI) Camera interface (CSI) Equipped with a charging chip and a battery detection chip Integrated microphone and speaker interfaces Micro SD Onboard 32GB eMMC storage and 4GB LPDDR4 memory Supports OTG functionality Equipped with an SMA antenna Dual Type-C ports User buttons Dual Grove ports 6-axis gyroscope Development platform StackFlow Specifications Processor AX630C @ Dual Cortex A53 1.2 GHzMAX.12.8 TOPs @INT4, and 3.2 TOPs @INT8 NPU 3.2 TOPs @ INT8 RAM 4GB LPDDR4 (2GB system memory + 2GB dedicated for hardware acceleration) eMMC eMMC5.1 @ 32GB Wired Network JL2101B-N040C @ 1GbE Wireless Network ESP32-C6 @ Wi-Fi6 2.4G USB-UART CH9102F @ USB to Serial Port USB-OTG USB 2.0 Host or Device Antenna Port SMA inner hole Audio Interface MIC and SPK Header 5P @ 1.25mm Display Interface MIPI DSI 1x 2Lane, MAX 1080p @ 30fps @ 1.25mm Camera Interface MIPI CSI 1x 4Lane, MAX 4K @ 30fps @ 1.25mm Battery Management AW32001ECSR charging chip, BQ27220YZFR battery detection chip Battery Interface Terminal 1.25mm spec battery interface terminal Compatible Battery Spec 3.7V lithium battery (Li-ion or Li-polymer) USB Ports 2 Type-C ports (data transfer, OTG function) USB Power Input 5V @ 2A Grove Interface PortA Header 4P @ 2.0mm (I2C)PortC Header 4P @ 2.0mm (UART) Storage Expansion Interface Micro SD card slot External Function Interface FUNC Header 8P @ 1.25mm (system wake, power management, external LED control, I2C, etc.) Buttons 2 buttons for power on/off, user interaction, reset, and other functions Sensor BMI270 @ 6-axis Operating Temperature 0C – 40C Product Dimensions 90.3 * 31.6 * 12mm Package Dimensions 133.0 * 83.0 * 17.0mm Product Weight 22.9g Package Weight 60.0g Resources Product documentation Package Contents 1x LLM630 Compute Kit 1x SMA 2.4G Wi-Fi Antenna 1x Heatsink 4x EVA Foam Foot Pads