--- product_id: 529380776 title: "SOM System-On-Modules - SOM Google Edge TPU ML Compute Accelerator, Integrate The Edge TPU into Legacy and New Systems Using a Standard Half-Mini PCIe" brand: "google coral" price: "788 zł" currency: PLN in_stock: true reviews_count: 5 category: "Google Coral" url: https://www.desertcart.pl/products/529380776-som-system-on-modules-som-google-edge-tpu-ml-compute store_origin: PL region: Poland --- # 4 TOPS per watt efficiency Supports TensorFlow Lite & AutoML Half-mini PCIe plug & play SOM System-On-Modules - SOM Google Edge TPU ML Compute Accelerator, Integrate The Edge TPU into Legacy and New Systems Using a Standard Half-Mini PCIe **Brand:** google coral **Price:** 788 zł **Availability:** ✅ In Stock ## Summary > 🤖 Power your edge AI with Google Coral’s lightning-fast TPU accelerator! ## Quick Answers - **What is this?** SOM System-On-Modules - SOM Google Edge TPU ML Compute Accelerator, Integrate The Edge TPU into Legacy and New Systems Using a Standard Half-Mini PCIe by google coral - **How much does it cost?** 788 zł with free shipping - **Is it available?** Yes, in stock and ready to ship - **Where can I buy it?** [www.desertcart.pl](https://www.desertcart.pl/products/529380776-som-system-on-modules-som-google-edge-tpu-ml-compute) ## Best For - google coral enthusiasts ## Why This Product - Trusted google coral brand quality - Free international shipping included - Worldwide delivery with tracking - 15-day hassle-free returns ## Key Features - • **Ready for AI at the Edge:** Run state-of-the-art MobileNet v2 models at 400 FPS for real-time vision tasks - • **Compact & Power Efficient:** 27mm footprint with DDR4 memory and ARM 4GHz processor for sleek, scalable edge AI - • **Blazing Fast ML Inferencing:** 4 trillion operations per second with ultra-low 0.5W per TOPS power draw - • **Seamless Debian Linux Integration:** Plug into any Debian-based system with standard half-mini PCIe slot—legacy or new - • **TensorFlow Lite & AutoML Compatible:** Deploy custom, high-accuracy models instantly without rebuilding from scratch ## Overview The SOM Google Edge TPU ML Compute Accelerator is a compact, power-efficient module designed to turbocharge machine learning inferencing at the edge. Delivering 4 trillion operations per second at just 0.5 watts per TOPS, it supports Debian-based Linux systems via a standard half-mini PCIe interface. Compatible with TensorFlow Lite and AutoML Vision Edge, it enables rapid deployment of custom AI models, making it ideal for professionals seeking cutting-edge, scalable AI acceleration in legacy or new hardware. ## Description Performs high-speed ML inferencing The on-board Edge TPU coprocessor is capable of performing 4 trillion operations (tera-operations) per second (TOPS), using 0.5 watts for each TOPS (2 TOPS per watt). For example, it can execute state-of-the-art mobile vision models such as MobileNet v2 at 400 FPS, in a power efficient manner. See more performance benchmarks. Works with Debian Linux Integrates with any Debian-based Linux system with a compatible card module slot. Supports TensorFlow Lite No need to build models from the ground up. TensorFlow Lite models can be compiled to run on the Edge TPU. Supports AutoML Vision Edge Easily build and deploy fast, high-accuracy custom image classification models to your device with AutoML Vision Edge. Review: Works on my Intel NUC - This works great on an old Intel NUC that I use to run Frigate NVR. Review: Doesn't work with most adaptors or computers. - I eventually got it working but it cost me a small fortune buying several expensive ($80+) adaptors and in the end only one PCIe adaptor would work in a Lenovo machine I have. Nothing would work in any of the Dell computers I have. The drivers don't work with new Linux distributions. I expected so much better from Google. Google isn't what they used to be, only in it for the money now. Sad. ## Features - 64-bit version of Debian 10 or Ubuntu 16.04 (or newer) - x86-64 or ARMv8 system architecture - 64-bit version of Windows 10 - x86-64 system architecture ## Technical Specifications | Specification | Value | |---------------|-------| | ASIN | B0844WRL58 | | Best Sellers Rank | #1,415,689 in Electronics ( See Top 100 in Electronics ) #26,387 in Components | | Box Contents | SOM Google Edge TPU ML Compute Accelerator board | | Brand | Google Coral | | Brand Name | Google Coral | | CPU Speed | 4 GHz | | Compatible Devices | Debian-based Linux systems | | Connectivity Technology | Half-size Mini PCIe | | Customer Reviews | 4.5 out of 5 stars 21 Reviews | | Included Components | SOM Google Edge TPU ML Compute Accelerator board | | Item Dimensions L x W x H | 27W Millimeters | | Manufacturer | Google Coral | | Manufacturer Part Number | G650-04528-01 | | Memory Storage Capacity | 8 GB | | Model Name | Coral SOM | | Model Number | G650-04528-01 | | Network Connectivity Technology | Half-size Mini PCIe | | Operating System | Debian-based Linux | | Processor Brand | ARM | | Processor Count | 1 | | Processor Speed | 4 GHz | | RAM Memory Technology | DDR4 | | UPC | 608614202775 | | Unit Count | 1 Count | | Wireless Compability | Bluetooth | ## Product Details - **Brand:** Google Coral - **Model Name:** Coral SOM - **Memory Storage Capacity:** 8 GB - **CPU Speed:** 4 GHz - **Connectivity Technology:** Half-size Mini PCIe ## Images ![SOM System-On-Modules - SOM Google Edge TPU ML Compute Accelerator, Integrate The Edge TPU into Legacy and New Systems Using a Standard Half-Mini PCIe - Image 1](https://m.media-amazon.com/images/I/51ggonhJIbL.jpg) ## Customer Reviews ### ⭐⭐⭐⭐⭐ Works on my Intel NUC *by B***. on 23 January 2025* This works great on an old Intel NUC that I use to run Frigate NVR. ### ⭐ Doesn't work with most adaptors or computers. *by G***D on 22 January 2024* I eventually got it working but it cost me a small fortune buying several expensive ($80+) adaptors and in the end only one PCIe adaptor would work in a Lenovo machine I have. Nothing would work in any of the Dell computers I have. The drivers don't work with new Linux distributions. I expected so much better from Google. Google isn't what they used to be, only in it for the money now. Sad. ### ⭐⭐⭐⭐ Works, not very reliable. You will have issues with drivers. *by H***G on 8 August 2025* Works, not very reliable. You will have issues with drivers. ## Frequently Bought Together - SOM System-On-Modules - SOM Google Edge TPU ML Compute Accelerator, Integrate The Edge TPU into Legacy and New Systems Using a Standard Half-Mini PCIe - Ableconn PEX-MP117 Mini PCI-E to PCI-E Adapter Card - Mini PCIe to PCIe Adapter - mPCIe to PCIe - Ideal for Wireless Mini PCI Express Card - GLOTRENDS WA02 Mini PCIe to PCIe X1 WiFi Bluetooth Adapter for Mini PCIe WiFi 4/5/6/6e Module/Coral TPU Module, with 3.5-dBi SMA Antenna Kits --- ## Why Shop on Desertcart? - 🛒 **Trusted by 1.3+ Million Shoppers** — Serving international shoppers since 2016 - 🌍 **Shop Globally** — Access 737+ million products across 21 categories - 💰 **No Hidden Fees** — All customs, duties, and taxes included in the price - 🔄 **15-Day Free Returns** — Hassle-free returns (30 days for PRO members) - 🔒 **Secure Payments** — Trusted payment options with buyer protection - ⭐ **TrustPilot Rated 4.5/5** — Based on 8,000+ happy customer reviews **Shop now:** [https://www.desertcart.pl/products/529380776-som-system-on-modules-som-google-edge-tpu-ml-compute](https://www.desertcart.pl/products/529380776-som-system-on-modules-som-google-edge-tpu-ml-compute) --- *Product available on Desertcart Poland* *Store origin: PL* *Last updated: 2026-09-24*