{"title":"AI \/ Boards \/ Kits \/ Raspberry Pi","description":null,"products":[{"product_id":"hailo-8-m-2-ai-accelerator-module-based-on-the-26tops-hailo-8-ai-processor-supports-linux-windows-systems","title":"Hailo-8 M.2 AI Accelerator Module, based on the 26TOPS Hailo-8 AI processor, Supports Linux\/Windows Systems","description":"\u003cp\u003e\u003cstrong\u003eKurzbeschreibung:\u003c\/strong\u003e Hailo-8 M.2 AI Accelerator Module, based on the 26TOPS Hailo-8 AI processor, Supports Linux\/Windows Systems | Hailo-8\u003c\/p\u003e\u003cdiv class=\"overviewSec\"\u003eHailo-8 M.2 AI Accelerator Module, based on the 26TOPS Hailo-8 AI processor, Supports Linux\/Windows Systems\u003c\/div\u003e\u003cdiv class=\"richWrap\"\u003e\n\u003cdiv class=\"pdT\" style=\"background: #dce6f0;\"\u003e\n\u003cdiv class=\"richTitle\"\u003eHailo-8 AI Accelerator Module\u003c\/div\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-1.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module, front view\"\u003e\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-3.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module, features\"\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\"\u003e\n\u003cdiv class=\"richTitle\"\u003eHailo-8 AI M.2 Module Parameters\u003c\/div\u003e\n\u003ctable class=\"tabSty-2 mgnTB alignl\"\u003e\u003ctbody\u003e\n\u003ctr\u003e\n\u003cth\u003eAI performance\u003c\/th\u003e\n\u003ctd\u003e26 TOPS\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eForm Factor\u003c\/th\u003e\n\u003ctd\u003eM.2 Key M\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003ePower supply\u003c\/th\u003e\n\u003ctd\u003e3.3V ± 5%\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003ePower consumption\u003c\/th\u003e\n\u003ctd\u003e2.5W (Typ.)  8.65W (Max.)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eInterface\u003c\/th\u003e\n\u003ctd\u003ePCIe Gen3, 4-lane\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eCertificate\u003c\/th\u003e\n\u003ctd\u003eCE, FCC Class A\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eStorage temperature\u003c\/th\u003e\n\u003ctd\u003e-40 ~ 85°C\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eOperating temperature\u003c\/th\u003e\n\u003ctd\u003e-40 ~ 85°C\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eOperating humidity\u003c\/th\u003e\n\u003ctd\u003e5% ~ 90%RH (no frosting)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eDimensions\u003c\/th\u003e\n\u003ctd\u003e22x42 (mm)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\u003c\/table\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\" style=\"background: #f5f5f5;\"\u003e\n\u003cdiv class=\"richTitle pdTRBL\"\u003eEquipped with Hailo-8 AI Accelerator to Step Up Your Edge Product Performance\u003c\/div\u003e\n\u003cp class=\"pdTRBL2 alignl\"\u003e   The Hailo-8 M.2 module is an AI accelerator module for AI applications, based on the 26 tera-operations per second (TOPS) Hailo-8 AI processor with high power efficiency. The M.2 AI accelerator features a full PCIe Gen-3.0 4-lane interface, delivering unprecedented AI performance for edge devices.       The M.2 module can be plugged into an existing edge device with M.2 socket to provide low-power deep neural network inferencing. Leveraging Hailo's comprehensive Dataflow Compiler and its support for standard AI frameworks, customers can easily port their Neural Network models to the Hailo-8 and introduce high-performance AI products to the market quickly.\u003c\/p\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-5.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module, equipped with Hailo-8 AI Accelerator\"\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\"\u003e\n\u003cdiv class=\"richTitle\"\u003eHailo-8 performance parameters\u003c\/div\u003e\n\u003ctable class=\"tabSty-2 mgnTB alignc\"\u003e\u003ctbody\u003e\n\u003ctr\u003e\n\u003cth\u003eNN Model\u003c\/th\u003e\n\u003cth\u003emAP\u003c\/th\u003e\n\u003cth\u003eHailo-8L FPS(batch8)\u003c\/th\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolov4_tiny\u003c\/td\u003e\n\u003ctd\u003e18.98\u003c\/td\u003e\n\u003ctd\u003e610\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolov6n\u003c\/td\u003e\n\u003ctd\u003e34.3\u003c\/td\u003e\n\u003ctd\u003e345\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolov7\u003c\/td\u003e\n\u003ctd\u003e49.8\u003c\/td\u003e\n\u003ctd\u003e45\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolox_s_wide\u003c\/td\u003e\n\u003ctd\u003e42.4\u003c\/td\u003e\n\u003ctd\u003e75\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolov3\u003c\/td\u003e\n\u003ctd\u003e38\u003c\/td\u003e\n\u003ctd\u003e26\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolov8n\u003c\/td\u003e\n\u003ctd\u003e37.23\u003c\/td\u003e\n\u003ctd\u003e270\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolov8s\u003c\/td\u003e\n\u003ctd\u003e44.75\u003c\/td\u003e\n\u003ctd\u003e128\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolov8m\u003c\/td\u003e\n\u003ctd\u003e50.08\u003c\/td\u003e\n\u003ctd\u003e55\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\u003c\/table\u003e\n\u003ctable class=\"tabSty-2 mgnTB alignc\"\u003e\u003ctbody\u003e\n\u003ctr\u003e\n\u003cth style=\"width: 20%;\"\u003eType\u003c\/th\u003e\n\u003cth\u003eNN Model\u003c\/th\u003e\n\u003cth\u003eInput Resolution\u003c\/th\u003e\n\u003cth\u003eFPS\u003c\/th\u003e\n\u003cth\u003ePower(W)\u003c\/th\u003e\n\u003cth\u003eFPS\/W\u003c\/th\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth rowspan=\"3\"\u003eClassification\u003c\/th\u003e\n\u003ctd\u003eResNet-50 v1\u003c\/td\u003e\n\u003ctd\u003e224x224\u003c\/td\u003e\n\u003ctd\u003e1332\u003c\/td\u003e\n\u003ctd\u003e3.45\u003c\/td\u003e\n\u003ctd\u003e386\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eMobileNet_v2_1.0\u003c\/td\u003e\n\u003ctd\u003e224x224\u003c\/td\u003e\n\u003ctd\u003e2444\u003c\/td\u003e\n\u003ctd\u003e2.152\u003c\/td\u003e\n\u003ctd\u003e1135\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eEfficientNet_M\u003c\/td\u003e\n\u003ctd\u003e240x240\u003c\/td\u003e\n\u003ctd\u003e889\u003c\/td\u003e\n\u003ctd\u003e3.5\u003c\/td\u003e\n\u003ctd\u003e254\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth rowspan=\"2\"\u003eObject Detection\u003c\/th\u003e\n\u003ctd\u003eSSD_MobileNet_v1\u003c\/td\u003e\n\u003ctd\u003e300x300\u003c\/td\u003e\n\u003ctd\u003e1055\u003c\/td\u003e\n\u003ctd\u003e2.2\u003c\/td\u003e\n\u003ctd\u003e479\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eYOLOv5m\u003c\/td\u003e\n\u003ctd\u003e640x640\u003c\/td\u003e\n\u003ctd\u003e218\u003c\/td\u003e\n\u003ctd\u003e4.6\u003c\/td\u003e\n\u003ctd\u003e47.3\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eSegmentation\u003c\/th\u003e\n\u003ctd\u003estdc1\u003c\/td\u003e\n\u003ctd\u003e1024x1920\u003c\/td\u003e\n\u003ctd\u003e54\u003c\/td\u003e\n\u003ctd\u003e2.9\u003c\/td\u003e\n\u003ctd\u003e18.6\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eMulti stream object detection (8 streams)\u003c\/th\u003e\n\u003ctd\u003eYOLOv3\u003c\/td\u003e\n\u003ctd\u003e608x608\u003c\/td\u003e\n\u003ctd\u003e69\u003c\/td\u003e\n\u003ctd\u003e4.9\u003c\/td\u003e\n\u003ctd\u003e14\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\u003c\/table\u003e\n\u003c\/div\u003e\n\u003c!--   \u003cdiv class=\"pdT\" style=\"background: #f0f7ff;\"\u003e\u003cdiv class=\"richTitle\"\u003eConnecting to PI5 via 16PIN Cable\u003c\/div\u003e\u003cp class=\"richDesc\"\u003eBased on 16PIN PCIe Interface of Raspberry Pi 5\u003c\/p\u003e\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-7.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module with PCIe TO M.2 HAT+, connecting to PI5 via 16PIN cable\" \/\u003e\u003cdiv\u003e\u003c\/div\u003e\u003c\/div\u003e --\u003e\u003c!-- \u003cdiv class=\"pdT\"\u003e\u003cdiv class=\"richTitle\"\u003etailored for Raspberry Pi 5\u003c\/div\u003e\u003cp class=\"richDesc pdTRBL\"\u003eStandard Raspberry Pi 40PIN Header, Comes with 2*20 Pin header for Stacking with Other HATs. Compact Size, More Space-Saving, supports installing cooling fan\u003c\/p\u003e\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-9.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module with PCIe TO M.2 HAT+, tailored for Raspberry Pi 5\" \/\u003e\u003c\/div\u003e --\u003e\u003c!--   \u003cdiv class=\"pdT\"\u003e\u003cdiv class=\"richTitle\"\u003eReserved airflow vent for Cooling Fan\u003c\/div\u003e\u003cp class=\"richDesc pdTRBL\"\u003ecan be used together with the \u003ca href=\"https:\/\/www.waveshare.com\/pi5-active-cooler-b.htm\" target=\"_blank\"\u003ePi5 Active Cooler B\u003c\/a\u003e to achieve better heat dissipation effect For the Pi5 And AI Accelerator Module, keeping it cool even under heavy processing and maximizing the Module Performance\u003c\/p\u003e\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-13.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module with PCIe TO M.2 HAT+, reserved airflow vent for Cooling Fan\" \/\u003e\u003c\/div\u003e--\u003e\u003c!--   \u003cdiv class=\"pdT\" style=\"background: #f0f7ff;\"\u003e\u003cdiv class=\"richTitle\"\u003eOnboard Power Monitoring Chip And EEPROM\u003c\/div\u003e\u003cp class=\"richDesc\"\u003eReal-time monitoring of device power status for More Stable operation\u003c\/p\u003e\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-11.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module with PCIe TO M.2 HAT+, onboard power monitoring chip and EEPROM\" \/\u003e\u003cdiv\u003e\u003c\/div\u003e\u003c\/div\u003e--\u003e\u003c!--   \u003cdiv class=\"pdT\"\u003e\u003cdiv class=\"richTitle\"\u003eHow to Install\u003c\/div\u003e\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-15.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module with PCIe TO M.2 HAT+, how to Install\" \/\u003e\u003cp class=\"note\"\u003e* for reference only, the cooling fan is NOT included.\u003c\/p\u003e\u003c\/div\u003e --\u003e\u003cdiv class=\"pdT\" style=\"background: #e9e9e9;\"\u003e\n\u003cdiv class=\"richTitle\"\u003eApplied in the AI Field\u003c\/div\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-17.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module, applied in the AI field\"\u003e\u003cdiv\u003e\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\"\u003e\n\u003cdiv class=\"richTitle\"\u003eOutline Dimensions\u003c\/div\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-size.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module and PCIe TO M.2 HAT+ dimensions\"\u003e\n\u003c\/div\u003e\n\u003cdiv\u003e\n\u003cdiv class=\"richTitle\"\u003eResources \u0026amp; Services\u003c\/div\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/LCD\/common\/res_banner.png\"\u003e\u003cdiv class=\"bgGrey\"\u003e\n\u003cstrong\u003eWiki:\u003c\/strong\u003e\u003ca href=\"http:\/\/www.waveshare.com\/wiki\/Hailo-8\" target=\"_blank\"\u003ewww.waveshare.com\/wiki\/Hailo-8\u003c\/a\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\u003ch2\u003eLieferumfang\u003c\/h2\u003e\u003cp\u003eHailo-8 AI M.2 Module x1\u003c\/p\u003e\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-pack.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module\"\u003e\u003cdiv class=\"sep0px\"\u003e\u003c\/div\u003e","brand":"HAILO","offers":[{"title":"Default Title","offer_id":53568484475217,"sku":"WSAL27812","price":81269.54,"currency_code":"HUF","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0983\/3798\/0753\/files\/wsal278121-459243.jpg?v=1776679258"},{"product_id":"hailo-8-acce-a-based-on-the-26tops-hailo-8-ai-processor-supports-linux-windows-systems-suitable-for-raspberry-pi-5","title":"Hailo-8 Acce A, based on the 26TOPS Hailo-8 AI processor, Supports Linux\/Windows Systems, Suitable For Raspberry Pi 5","description":"\u003cp\u003e\u003cstrong\u003eKurzbeschreibung:\u003c\/strong\u003e Hailo-8 Acce A, based on the 26TOPS Hailo-8 AI processor, Supports Linux\/Windows Systems, Suitable For Raspberry Pi 5\u003c\/p\u003e\u003cdiv class=\"richWrap\"\u003e\n\u003cdiv class=\"pdT\" style=\"background: #dce6f0;\"\u003e\n\u003cdiv class=\"richTitle\"\u003eHailo-8 AI Kit\u003c\/div\u003e\n\u003cp class=\"richDesc hilitBlock3 round1234\" style=\"background: #2d9fe8; color: #FFFFFF;\"\u003e Equipped With 26TOPS Hailo-8 M.2 AI Accelerator Module\u003c\/p\u003e\n\u003cp class=\"pdTRBL2 alignl\"\u003eThis AI kit is launched by Waveshare to provide a more cost-effective and high-performance AI solution for the Raspberry Pi 5, suitable for applications such as process control, safety, home automation and robotics, etc.\u003c\/p\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-Acce-A-details-1.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module, front view\"\u003e\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-3.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module, features\"\u003e\u003c!-- \u003cdiv class=\"pdT\" style=\"background: #f8f8f8;\"\u003e\u003cdiv class=\"richTitle\"\u003eVersion Options\u003c\/div\u003e\u003cdiv class=\"richGridWrap richGridWrap-2c richGridWrap-valignTop mgnTB\"\u003e\u003cdiv\u003e\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-1_420.jpg\" class=\"round1234\" alt=\"Hailo-8, front view\" \/\u003e\u003cp\u003e\u003cstrong\u003eHailo-8\u003c\/strong\u003ethe Hailo-8 AI M.2 module only\u003c\/p\u003e\u003c\/div\u003e\u003cdiv\u003e\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-3-2.jpg\" class=\"round1234\"  alt=\"Hailo-8 Acce A, front view\" \/\u003e\u003cp\u003e\u003cstrong\u003eHailo-8 Acce A\u003c\/strong\u003eHailo-8 M.2 module + PCIe TO M.2 adapter and accessories (can be directly accessed to Raspberry Pi 5)\u003c\/p\u003e\u003c\/div\u003e\u003c\/div\u003e\u003c\/div\u003e --\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\"\u003e\n\u003cdiv class=\"richTitle\"\u003eFeatures At A Glance\u003c\/div\u003e\n\u003cdiv class=\"pdTRBL2 alignl\"\u003e\u003cul\u003e\n\u003cli\u003eHailo-8 AI M.2 module  \u003cul\u003e\n\u003cli\u003ePowered by 26 Tera-Operations Per Second (TOPS) Hailo-8 AI Processor\u003c\/li\u003e\n\u003cli\u003e2.5W typical power consumption\u003c\/li\u003e\n\u003cli\u003eScalable,enabling simultaneous processingof multi-streams \u0026amp; multi-models\u003c\/li\u003e\n\u003cli\u003eEnabling real-time, low latency and high-efficiency AI inferencing on the edge devices\u003c\/li\u003e\n\u003cli\u003eSupports TensorFlow, TensorFlow Lite, ONNX, Keras, Pytorch frameworks\u003c\/li\u003e\n\u003cli\u003eSupports Linux and Windows\u003c\/li\u003e\n\u003cli\u003eSupports the temperature range of -40°C to 85°C\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli\u003ePCIe To M.2 adapter  \u003cul\u003e\n\u003cli\u003eOnboard power monitoring chip and EEPROM, supports real-time monitoring of device power status for more stable operation\u003c\/li\u003e\n\u003cli\u003eRaspberry Pi HAT+ compliant\u003c\/li\u003e\n\u003cli\u003eReserved airflow vent, supports installing cooling fan for better heat dissipation of the AI module to improve performance\u003c\/li\u003e\n\u003cli\u003eImmersion gold process design, anti-oxidation and more durable\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\"\u003e\n\u003cdiv class=\"richTitle\"\u003eHailo-8 AI M.2 Module Parameters\u003c\/div\u003e\n\u003ctable class=\"tabSty-2 mgnTB alignl\"\u003e\u003ctbody\u003e\n\u003ctr\u003e\n\u003cth\u003eAI performance\u003c\/th\u003e\n\u003ctd\u003e26 TOPS\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eForm Factor\u003c\/th\u003e\n\u003ctd\u003eM.2 Key M\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003ePower supply\u003c\/th\u003e\n\u003ctd\u003e3.3V ± 5%\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003ePower consumption\u003c\/th\u003e\n\u003ctd\u003e2.5W (Typ.)  8.65W (Max.)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eInterface\u003c\/th\u003e\n\u003ctd\u003ePCIe Gen3, 4-lane\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eCertificate\u003c\/th\u003e\n\u003ctd\u003eCE, FCC Class A\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eStorage temperature\u003c\/th\u003e\n\u003ctd\u003e-40 ~ 85°C\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eOperating temperature\u003c\/th\u003e\n\u003ctd\u003e-40 ~ 85°C\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eOperating humidity\u003c\/th\u003e\n\u003ctd\u003e5% ~ 90%RH (no frosting)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eDimensions\u003c\/th\u003e\n\u003ctd\u003e22×80mm with breakable extensions to22×42mmand 22×60mm\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\u003c\/table\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\" style=\"background: #f5f5f5;\"\u003e\n\u003cdiv class=\"richTitle pdTRBL\"\u003eEquipped with Hailo-8 AI Accelerator to Step Up Your Edge Product Performance\u003c\/div\u003e\n\u003cp class=\"pdTRBL2 alignl\"\u003e   The Hailo-8 M.2 module is an AI accelerator module for AI applications, based on the 26 tera-operations per second (TOPS) Hailo-8 AI processor with high power efficiency. The M.2 AI accelerator features a full PCIe Gen-3.0 4-lane interface, delivering unprecedented AI performance for edge devices.       The M.2 module can be plugged into an existing edge device with M.2 socket to provide low-power deep neural network inferencing. Leveraging Hailo's comprehensive Dataflow Compiler and its support for standard AI frameworks, customers can easily port their Neural Network models to the Hailo-8 and introduce high-performance AI products to the market quickly.\u003c\/p\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-5.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module, equipped with Hailo-8 AI Accelerator\"\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\"\u003e\n\u003cdiv class=\"richTitle\"\u003eHailo-8 performance parameters\u003c\/div\u003e\n\u003ctable class=\"tabSty-2 mgnTB alignc\"\u003e\u003ctbody\u003e\n\u003ctr\u003e\n\u003cth\u003eNN Model\u003c\/th\u003e\n\u003cth\u003emAP\u003c\/th\u003e\n\u003cth\u003eHailo-8L FPS(batch8)\u003c\/th\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolov4_tiny\u003c\/td\u003e\n\u003ctd\u003e18.98\u003c\/td\u003e\n\u003ctd\u003e610\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolov6n\u003c\/td\u003e\n\u003ctd\u003e34.3\u003c\/td\u003e\n\u003ctd\u003e345\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolov7\u003c\/td\u003e\n\u003ctd\u003e49.8\u003c\/td\u003e\n\u003ctd\u003e45\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolox_s_wide\u003c\/td\u003e\n\u003ctd\u003e42.4\u003c\/td\u003e\n\u003ctd\u003e75\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolov3\u003c\/td\u003e\n\u003ctd\u003e38\u003c\/td\u003e\n\u003ctd\u003e26\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolov8n\u003c\/td\u003e\n\u003ctd\u003e37.23\u003c\/td\u003e\n\u003ctd\u003e270\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolov8s\u003c\/td\u003e\n\u003ctd\u003e44.75\u003c\/td\u003e\n\u003ctd\u003e128\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eyolov8m\u003c\/td\u003e\n\u003ctd\u003e50.08\u003c\/td\u003e\n\u003ctd\u003e55\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\u003c\/table\u003e\n\u003ctable class=\"tabSty-2 mgnTB alignc\"\u003e\u003ctbody\u003e\n\u003ctr\u003e\n\u003cth style=\"width: 20%;\"\u003eType\u003c\/th\u003e\n\u003cth\u003eNN Model\u003c\/th\u003e\n\u003cth\u003eInput Resolution\u003c\/th\u003e\n\u003cth\u003eFPS\u003c\/th\u003e\n\u003cth\u003ePower(W)\u003c\/th\u003e\n\u003cth\u003eFPS\/W\u003c\/th\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth rowspan=\"3\"\u003eClassification\u003c\/th\u003e\n\u003ctd\u003eResNet-50 v1\u003c\/td\u003e\n\u003ctd\u003e224x224\u003c\/td\u003e\n\u003ctd\u003e1332\u003c\/td\u003e\n\u003ctd\u003e3.45\u003c\/td\u003e\n\u003ctd\u003e386\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eMobileNet_v2_1.0\u003c\/td\u003e\n\u003ctd\u003e224x224\u003c\/td\u003e\n\u003ctd\u003e2444\u003c\/td\u003e\n\u003ctd\u003e2.152\u003c\/td\u003e\n\u003ctd\u003e1135\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eEfficientNet_M\u003c\/td\u003e\n\u003ctd\u003e240x240\u003c\/td\u003e\n\u003ctd\u003e889\u003c\/td\u003e\n\u003ctd\u003e3.5\u003c\/td\u003e\n\u003ctd\u003e254\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth rowspan=\"2\"\u003eObject Detection\u003c\/th\u003e\n\u003ctd\u003eSSD_MobileNet_v1\u003c\/td\u003e\n\u003ctd\u003e300x300\u003c\/td\u003e\n\u003ctd\u003e1055\u003c\/td\u003e\n\u003ctd\u003e2.2\u003c\/td\u003e\n\u003ctd\u003e479\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eYOLOv5m\u003c\/td\u003e\n\u003ctd\u003e640x640\u003c\/td\u003e\n\u003ctd\u003e218\u003c\/td\u003e\n\u003ctd\u003e4.6\u003c\/td\u003e\n\u003ctd\u003e47.3\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eSegmentation\u003c\/th\u003e\n\u003ctd\u003estdc1\u003c\/td\u003e\n\u003ctd\u003e1024x1920\u003c\/td\u003e\n\u003ctd\u003e54\u003c\/td\u003e\n\u003ctd\u003e2.9\u003c\/td\u003e\n\u003ctd\u003e18.6\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eMulti stream object detection (8 streams)\u003c\/th\u003e\n\u003ctd\u003eYOLOv3\u003c\/td\u003e\n\u003ctd\u003e608x608\u003c\/td\u003e\n\u003ctd\u003e69\u003c\/td\u003e\n\u003ctd\u003e4.9\u003c\/td\u003e\n\u003ctd\u003e14\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\u003c\/table\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\" style=\"background: #f0f7ff;\"\u003e\n\u003cdiv class=\"richTitle\"\u003eConnecting to PI5 via 16PIN Cable\u003c\/div\u003e\n\u003cp class=\"richDesc\"\u003eBased on 16PIN PCIe Interface of Raspberry Pi 5\u003c\/p\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-7.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module with PCIe TO M.2 HAT+, connecting to PI5 via 16PIN cable\"\u003e\u003cdiv\u003e\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\"\u003e\n\u003cdiv class=\"richTitle\"\u003etailored for Raspberry Pi 5\u003c\/div\u003e\n\u003cp class=\"richDesc pdTRBL\"\u003eStandard Raspberry Pi 40PIN Header, Comes with 2*20 Pin header for Stacking with Other HATs. Compact Size, More Space-Saving, supports installing cooling fan\u003c\/p\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-9.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module with PCIe TO M.2 HAT+, tailored for Raspberry Pi 5\"\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\"\u003e\n\u003cdiv class=\"richTitle\"\u003eReserved airflow vent for Cooling Fan\u003c\/div\u003e\n\u003cp class=\"richDesc pdTRBL\"\u003ecan be used together with the \u003ca href=\"https:\/\/www.waveshare.com\/pi5-active-cooler-b.htm\" target=\"_blank\"\u003ePi5 Active Cooler B\u003c\/a\u003e to achieve better heat dissipation effect For the Pi5 And AI Accelerator Module, keeping it cool even under heavy processing and maximizing the Module Performance\u003c\/p\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-13.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module with PCIe TO M.2 HAT+, reserved airflow vent for Cooling Fan\"\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\" style=\"background: #f0f7ff;\"\u003e\n\u003cdiv class=\"richTitle\"\u003eOnboard Power Monitoring Chip And EEPROM\u003c\/div\u003e\n\u003cp class=\"richDesc\"\u003eReal-time monitoring of device power status for More Stable operation\u003c\/p\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-11.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module with PCIe TO M.2 HAT+, onboard power monitoring chip and EEPROM\"\u003e\u003cdiv\u003e\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\"\u003e\n\u003cdiv class=\"richTitle\"\u003eHow to Install\u003c\/div\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-15.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module with PCIe TO M.2 HAT+, how to Install\"\u003e\u003cp class=\"note\"\u003e* for reference only, the cooling fan is NOT included.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\" style=\"background: #e9e9e9;\"\u003e\n\u003cdiv class=\"richTitle\"\u003eApplied in the AI Field\u003c\/div\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-details-17.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module, applied in the AI field\"\u003e\u003cdiv\u003e\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\"\u003e\n\u003cdiv class=\"richTitle\"\u003eOutline Dimensions\u003c\/div\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-Acce-A-details-size.jpg\" alt=\"Hailo-8 M.2 AI Accelerator Module and PCIe TO M.2 HAT+ dimensions\"\u003e\n\u003c\/div\u003e\n\u003cdiv\u003e\n\u003cdiv class=\"richTitle\"\u003eResources \u0026amp; Services\u003c\/div\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/LCD\/common\/res_banner.png\"\u003e\u003cdiv class=\"bgGrey\"\u003e\n\u003cstrong\u003eWiki:\u003c\/strong\u003e\u003ca href=\"http:\/\/www.waveshare.com\/wiki\/Hailo-8\" target=\"_blank\"\u003ewww.waveshare.com\/wiki\/Hailo-8\u003c\/a\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\u003ch2\u003eLieferumfang\u003c\/h2\u003e\u003col\u003e\n\u003cli\u003eHailo-8 AI M.2 Module x1\u003c\/li\u003e\n\u003cli\u003ePCIe TO M.2 HAT+ x1\u003c\/li\u003e\n\u003cli\u003eStandoffs pack x1\u003c\/li\u003e\n\u003cli\u003e16P-Cable-40mm x1\u003c\/li\u003e\n\u003cli\u003e2*20 Pin header x1\u003c\/li\u003e\n\u003c\/ol\u003e\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Hailo-8\/Hailo-8-Acce-A-details-pack.jpg\" alt=\"Hailo-8 Acce A Kit\"\u003e","brand":"WaveShare","offers":[{"title":"Default Title","offer_id":53568484507985,"sku":"WSAL27841","price":84881.72,"currency_code":"HUF","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0983\/3798\/0753\/files\/wsal278411-459261.jpg?v=1776679276"},{"product_id":"official-raspberry-pi-ai-hat-2-built-in-hailo-10h-ai-accelerator-and-8gb-onboard-ram-comes-with-heatsink","title":"Official Raspberry Pi AI HAT+ 2, Built-in Hailo-10H AI Accelerator And 8GB Onboard RAM, Comes With Heatsink","description":"\u003cdiv class=\"msg msg_warning\"\u003e    Prohibited by the manufacturer Raspberry Pi, we're not allowed to ship this product, or to provide any tech support, to the following countries\/regions: Cuba, Iran, North Korea, Syria, Crimea, Donetsk, Luhansk, Russia, Belarus.  \u003c\/div\u003e\u003cdiv class=\"overviewSec\"\u003eOfficial Raspberry Pi AI HAT+ 2, Built-in Hailo-10H AI Accelerator and 8GB dedicated onboard RAM, brings generative AI capability to Raspberry Pi 5\u003c\/div\u003e\u003cdiv class=\"richWrap\"\u003e\n\u003cdiv class=\"pdT\"\u003e\n\u003cdiv class=\"richTitle\" style=\"color: #ce224e;\"\u003eRaspberry Pi AI HAT+ 2\u003c\/div\u003e\n\u003cp class=\"hilitBlock2 richDesc\" style=\"background: #ce224e;\"\u003e        High-performance AI HAT Designed By Raspberry Pi      \u003c\/p\u003e\n\u003cp class=\"richDesc\" style=\"color: #ce224e;\"\u003e        Built-in Hailo-10H AI Accelerator and 8GB dedicated onboard RAM, brings generative AI capability to Raspberry Pi 5      \u003c\/p\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Raspberry-Pi-AI-HAT-Plus-2\/Raspberry-Pi-AI-HAT-Plus-2-details-1.jpg\" alt=\"Raspberry Pi AI HAT+ 2, front view\"\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\"\u003e\n\u003cdiv class=\"richTitle\" style=\"color: #ce224e;\"\u003eSpecifications\u003c\/div\u003e\n\u003ctable class=\"tabSty-2 alignc mgnTB\"\u003e\u003ctbody\u003e\n\u003ctr\u003e\n\u003cth style=\"width: 30%;\"\u003eModel\u003c\/th\u003e\n\u003ctd style=\"width: 35%;\"\u003e\u003ca href=\"https:\/\/www.waveshare.com\/raspberry-pi-ai-hat-plus.htm\" target=\"_blank\" style=\"text-decoration: underline;\"\u003eRaspberry Pi AI HAT+\u003c\/a\u003e\u003c\/td\u003e\n\u003ctd style=\"width: 35%;\"\u003e\u003ca href=\"https:\/\/www.waveshare.com\/raspberry-pi-ai-hat-plus-2.htm\" target=\"_blank\" style=\"text-decoration: underline;\"\u003eRaspberry Pi AI HAT+ 2\u003c\/a\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eAccelerator chip (Hailo NPU)\u003c\/th\u003e\n\u003ctd\u003eHailo-8L (13 TOPS, INT8) \/  Hailo-8 (26 TOPS, INT8)\u003c\/td\u003e\n\u003ctd\u003eHailo-10H (40 TOPS, INT4)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eMemory (RAM)\u003c\/th\u003e\n\u003ctd\u003eUses the memory on a Raspberry Pi 5\u003c\/td\u003e\n\u003ctd\u003eHas its own 8 GB onboard memory, allowing it to run LLMs and VLMs up to ~6 billion parameters\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eLarge Language Model (LLM) support\u003c\/th\u003e\n\u003ctd\u003eNot supported\u003c\/td\u003e\n\u003ctd\u003eSupported\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eVision-Language Model (VLM) support\u003c\/th\u003e\n\u003ctd\u003eNot supported\u003c\/td\u003e\n\u003ctd\u003eSupported\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eUse cases\u003c\/th\u003e\n\u003ctd\u003eObject detection, camera post-processing, robotics, moderate neural workloads\u003c\/td\u003e\n\u003ctd\u003eEverything available on AI HAT+ plus generative AI workloads, including local LLMs and VLMs\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\u003c\/table\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\" style=\"background: #f5f5f5;\"\u003e\n\u003cdiv class=\"richTitle\" style=\"color: #ce224e;\"\u003eFeatures\u003c\/div\u003e\n\u003cdiv class=\"pdTRBL2 alignl\"\u003e\n\u003cp\u003eThe\u003cstrong\u003eRaspberry Pi AI HAT+ 2\u003c\/strong\u003e is an add-on board based on the Hailo-10H AI accelerator that brings generative AI capability to Raspberry Pi 5. With 8GB of dedicated on-board RAM, the AI HAT+ 2 is ideal for running large language models (LLMs) and vision-language models (VLMs) locally, leaving the host Raspberry Pi 5 free to handle other tasks. The AI HAT+ 2 delivers reliable, low-latency, accelerated generative AI at the edge, making it the perfect choice for applications including offline process control, secure data analysis, facilities management, and robotics.\u003c\/p\u003e\n\u003cp\u003eThe AI HAT+ 2 delivers 40 tera-operations per second (TOPS) of INT4 inferencing performance, with computer vision performance equivalent or superior to the 26 TOPS \u003ca href=\"https:\/\/www.waveshare.com\/raspberry-pi-ai-hat-plus.htm?sku=29050\" target=\"_blank\" style=\"text-decoration: underline;\"\u003eRaspberry Pi AI HAT+\u003c\/a\u003e. A set of sample models is provided by Hailo; users can also train custom vision models or fine-tune generative AI models using Low-Rank Adaptation (LoRA) to suit their application, such as speech to text, translation, or visual scene analysis.\u003c\/p\u003e\n\u003cp\u003eLike the original \u003ca href=\"https:\/\/www.waveshare.com\/raspberry-pi-ai-hat-plus.htm\" target=\"_blank\" style=\"text-decoration: underline;\"\u003eRaspberry Pi AI HAT+\u003c\/a\u003e, the AI HAT+ 2 communicates using Raspberry Pi 5’s PCI Express interface. When the host Raspberry Pi 5 is running an up-to-date Raspberry Pi OS image, it automatically detects the on-board Hailo accelerator and makes it available for AI computing tasks. The built-in rpicam-apps camera applications in Raspberry Pi OS natively support the AI module, automatically using the accelerator to run compatible post-processing tasks.\u003c\/p\u003e\n\u003cp style=\"color: #ce224e;\"\u003e\u003cstrong\u003eKey features include:\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul\u003e\n\u003cli\u003eHailo-10H AI accelerator delivering 40 TOPS (INT4) inferencing performance\u003c\/li\u003e\n\u003cli\u003ePerformance for computer vision models comparable to the \u003ca href=\"https:\/\/www.waveshare.com\/raspberry-pi-ai-hat-plus.htm?sku=29050\" target=\"_blank\" style=\"text-decoration: underline;\"\u003eRaspberry Pi AI HAT+ (26 TOPS)\u003c\/a\u003e\n\u003c\/li\u003e\n\u003cli\u003eRuns generative AI models efficiently using 8GB on-board RAM\u003c\/li\u003e\n\u003cli\u003eFully integrated into Raspberry Pi’s camera software stack\u003c\/li\u003e\n\u003cli\u003eConforms to Raspberry Pi HAT+ specification\u003c\/li\u003e\n\u003cli\u003eSupplied with heatsink, and with 16mm stacking header, spacers, and screws to enable fitting on Raspberry Pi 5 with \u003ca href=\"https:\/\/www.waveshare.com\/raspberry-pi-5-official-active-cooler.htm\" target=\"_blank\" style=\"text-decoration: underline;\"\u003eRaspberry Pi 5 Official Active Cooler\u003c\/a\u003e in place\u003c\/li\u003e\n\u003cli\u003eOperating temperature: 0°C ~ 50°C (ambient)\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/div\u003e\n\u003cdiv\u003e\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdTRBL2\" style=\"background: #ce224e;\"\u003e\n\u003cdiv class=\"richTitle\" style=\"color: #ffffff;\"\u003eOff‑board tasks with onboard RAM\u003c\/div\u003e\n\u003cdiv class=\"bgWhite\" style=\"margin-bottom: 0; border-radius: 0.5em 0.5em 0 0; padding: 0;\"\u003e\u003cimg style=\"border-radius: 0.5em 0.5em 0 0;\" src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Raspberry-Pi-AI-HAT-Plus-2\/Raspberry-Pi-AI-HAT-Plus-2-details-3.jpg\" alt=\"Raspberry Pi AI HAT+ 2, onboard RAM, standalone computing\"\u003e\u003c\/div\u003e\n\u003cdiv class=\"bgWhite\" style=\"background-color: #fbeef1; margin-top: 0;border-radius: 0 0 0.5em 0.5em; \"\u003e\u003cp class=\"alignl\" style=\"padding: 0 1em;\"\u003e          With 8GB of dedicated on-board RAM, the AI HAT+ 2 is ideal for running large language models (LLMs) and vision-language models (VLMs) locally, leaving the host Raspberry Pi 5 free to handle other tasks and ensuring your AI models run smoothly.        \u003c\/p\u003e\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdTRBL2\" style=\"background: #ce224e;\"\u003e\n\u003cdiv class=\"richTitle\" style=\"color: #ffffff;\"\u003eA model for privacy and security\u003c\/div\u003e\n\u003cdiv class=\"bgWhite\" style=\"margin-bottom: 0; border-radius: 0.5em 0.5em 0 0; padding: 0;\"\u003e\u003cimg style=\"border-radius: 0.5em 0.5em 0 0;\" src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Raspberry-Pi-AI-HAT-Plus-2\/Raspberry-Pi-AI-HAT-Plus-2-details-7.jpg\" alt=\"Raspberry Pi AI HAT+ 2, Ensuring user privacy and security\"\u003e\u003c\/div\u003e\n\u003cdiv class=\"bgWhite\" style=\"background-color: #fbeef1; margin-top: 0;border-radius: 0 0 0.5em 0.5em; \"\u003e\u003cp class=\"alignl\" style=\"padding: 0 1em;\"\u003e          Delivering reliable, low-latency AI processing at the edge and without a network connection, the AI HAT+ 2 simplifies data security, minimises infrastructure requirements, and reduces cloud API costs.        \u003c\/p\u003e\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\"\u003e\n\u003cdiv class=\"richTitle\" style=\"color: #ce224e;\"\u003ehow to install\u003c\/div\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Raspberry-Pi-AI-HAT-Plus-2\/Raspberry-Pi-AI-HAT-Plus-2-details-9.jpg\" alt=\"Raspberry Pi AI HAT+ 2, how to install\"\u003e\u003cp class=\"note\"\u003e* for reference only, the \u003ca href=\"https:\/\/www.waveshare.com\/raspberry-pi-5.htm\" target=\"_blank\"\u003eRaspberry Pi 5\u003c\/a\u003eand\u003ca href=\"https:\/\/www.waveshare.com\/raspberry-pi-5-official-active-cooler.htm\" target=\"_blank\" style=\"text-decoration: underline;\"\u003eRaspberry Pi 5 Official Active Cooler\u003c\/a\u003e  are NOT included.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\"\u003e\n\u003cdiv class=\"richTitle\" style=\"color: #ce224e;\"\u003eProduct show\u003c\/div\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Raspberry-Pi-AI-HAT-Plus-2\/Raspberry-Pi-AI-HAT-Plus-2-details-11.jpg\" alt=\"Raspberry Pi AI HAT+ 2, product show\"\u003e\u003cp class=\"note\"\u003e* for reference only, the \u003ca href=\"https:\/\/www.waveshare.com\/raspberry-pi-5.htm\" target=\"_blank\"\u003eRaspberry Pi 5\u003c\/a\u003eand\u003ca href=\"https:\/\/www.waveshare.com\/raspberry-pi-5-official-active-cooler.htm\" target=\"_blank\" style=\"text-decoration: underline;\"\u003eRaspberry Pi 5 Official Active Cooler\u003c\/a\u003e  are NOT included.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"pdT\"\u003e\n\u003cdiv class=\"richTitle\" style=\"color: #ce224e;\"\u003eoutline dimensions\u003c\/div\u003e\n\u003cp class=\"hilitBlock2\" style=\"background-color: #ce224e;\"\u003eRaspberry Pi AI HAT+ 2\u003c\/p\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Raspberry-Pi-AI-HAT-Plus-2\/Raspberry-Pi-AI-HAT-Plus-2-details-size.jpg\" alt=\"Raspberry Pi AI HAT+ 2, outline dimensions\"\u003e\u003cp class=\"hilitBlock2\" style=\"background-color: #ce224e;\"\u003eHeatsink\u003c\/p\u003e\n\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Raspberry-Pi-AI-HAT-Plus-2\/Raspberry-Pi-AI-HAT-Plus-2-details-size-1.jpg\" alt=\"Raspberry Pi AI HAT+ 2, Heatsink outline dimensions\"\u003e\n\u003c\/div\u003e\n\u003cdiv\u003e\n\u003cdiv class=\"richTitle\"\u003eResources \u0026amp; Services\u003c\/div\u003e\n\u003cdiv class=\"bgGrey\"\u003e\n\u003cstrong\u003eWiki:\u003c\/strong\u003e\u003ca href=\"http:\/\/www.waveshare.com\/wiki\/Raspberry_Pi_AI_HAT+_2\" target=\"_blank\"\u003ewww.waveshare.com\/wiki\/Raspberry_Pi_AI_HAT+_2\u003c\/a\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\u003ch2\u003eLieferumfang\u003c\/h2\u003e\u003col\u003e\n\u003cli\u003eRaspberry Pi AI HAT+ 2 x1\u003c\/li\u003e\n\u003cli\u003eRaspberry Pi 5 Official Heatsink x1\u003c\/li\u003e\n\u003cli\u003e16mm stacking header x1\u003c\/li\u003e\n\u003cli\u003eScrews pack x1\u003c\/li\u003e\n\u003c\/ol\u003e\u003cimg src=\"https:\/\/www.waveshare.com\/img\/devkit\/accBoard\/Raspberry-Pi-AI-HAT-Plus-2\/Raspberry-Pi-AI-HAT-Plus-2-details-pack.jpg\" alt=\"Raspberry Pi AI HAT+ 2\"\u003e","brand":"Raspberry Pi Foundation","offers":[{"title":"Default Title","offer_id":53568549355857,"sku":"WSAL33785","price":106554.83,"currency_code":"HUF","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0983\/3798\/0753\/files\/wsal337851-486003.jpg?v=1776699090"}],"url":"https:\/\/albistech.eu\/collections\/ai__boards_kits__raspberry_pi.oembed","provider":"Albistech","version":"1.0","type":"link"}