--- title: Intel's cheapest CPUs do what Google's discontinued... description: Google's Coral AI accelerators were once the go-to hardware for running local AI inference in self-hosted NVR setups like Frigate, but Google discontinued the... canonical: https://daily.dev/posts/intel-s-cheapest-cpus-do-what-google-s-discontinued-coral-accelerators-used-to-do-and-they-cost-les-xrc1swqbm twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Intel's cheapest CPUs do what Google's discontinued Coral accelerators used to do, and they cost less | daily.dev og:description: Google's Coral AI accelerators were once the go-to hardware for running local AI inference in self-hosted NVR setups like Frigate, but Google discontinued the... og:url: https://daily.dev/posts/intel-s-cheapest-cpus-do-what-google-s-discontinued-coral-accelerators-used-to-do-and-they-cost-les-xrc1swqbm og:image: https://api.daily.dev/og/posts/xrC1SWQbm.png og:image:alt: Intel's cheapest CPUs do what Google's discontinued Coral accelerators used to do, and they cost less og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Intel's cheapest CPUs do what Google's discontinued Coral accelerators used to do, and they cost less **[XDA Developers](https://daily.dev/sources/xda-developers)** · 4 min read · 5 upvotes · 0 comments ## Summary Google's Coral AI accelerators were once the go-to hardware for running local AI inference in self-hosted NVR setups like Frigate, but Google discontinued the product and its libraries haven't been updated since 2022. Frigate 0.17 has stopped recommending Coral and now steers GPU-less users toward Intel OpenVINO, which runs on budget Intel N100 mini-PCs without needing a dedicated accelerator. For SBC users, the Raspberry Pi AI Kit with Hailo-8L is a viable alternative, though the author recommends an Intel N100 mini-PC with OpenVINO running YOLOv9/MobileNet v2 as the most practical and cost-effective setup for local security camera object detection. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### Why did Frigate stop recommending Google Coral TPU accelerators? Frigate 0.17 stopped recommending Coral accelerators because Google discontinued support for the Coral TPU lineup and most required libraries have not been updated since 2022. The GASKET driver needed for PCI-based Coral devices is not supported on modern Linux kernels, requiring workarounds. Existing Coral setups still work, but Frigate now steers GPU-less users toward Intel OpenVINO instead. _Anyone rebuilding a self-hosted NVR pipeline can track shifts like this Coral-to-OpenVINO move on daily.dev._ ### What can I use instead of a Google Coral accelerator for Frigate object detection? Intel OpenVINO running on an inexpensive CPU like the Intel N100, or on an Intel NPU in a mini-PC, can replace a Coral accelerator for Frigate object detection, running models such as Yolov9 or MobileNet v2. It may draw slightly more power than a dedicated Coral Edge TPU, but avoids buying discontinued or unsupported hardware. _Comparing self-hosted AI accelerator options gets easier when the alternatives are tracked in one place on daily.dev._ ### Can a Raspberry Pi 5 handle AI-based object detection for security cameras on its own? A Raspberry Pi 5 can run AI detection tasks on its own CPU, but doing so requires lowering stream resolution and configuring large masks to keep CPU utilization from maxing out. Adding a Raspberry Pi AI Kit with a Hailo-8L accelerator offloads this inference work, allowing full-resolution camera streams without straining the SBC's processor. _Developers weighing SBC hardware add-ons for local AI workloads can follow updates like this on daily.dev._ ## Similar posts on daily.dev - [I bought a Coral accelerator for Frigate and couldn't get it to work — turns out I didn't need it anyway](https://daily.dev/posts/i-bought-a-coral-accelerator-for-frigate-and-couldn-t-get-it-to-work-turns-out-i-didn-t-need-it-an-bd7prprab) · XDA Developers · 2 upvotes · 0 comments - [Frigate with Hailo for object detection on a Raspberry Pi](https://daily.dev/posts/frigate-with-hailo-for-object-detection-on-a-raspberry-pi-9dnvfihpz) · Jeff Geerling · 0 upvotes · 0 comments - [Introducing Coral NPU: A full-stack platform for Edge AI](https://daily.dev/posts/introducing-coral-npu-a-full-stack-platform-for-edge-ai-dwohqdxtz) · Google Developers · 0 upvotes · 0 comments --- Tags: [#raspberry-pi](https://daily.dev/tags/raspberry-pi), [#object-detection](https://daily.dev/tags/object-detection) [View this post on daily.dev](https://daily.dev/posts/intel-s-cheapest-cpus-do-what-google-s-discontinued-coral-accelerators-used-to-do-and-they-cost-les-xrc1swqbm) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Intel's cheapest CPUs do what Google's discontinued Coral accelerators used to do, and they cost less","url":"https://daily.dev/posts/intel-s-cheapest-cpus-do-what-google-s-discontinued-coral-accelerators-used-to-do-and-they-cost-les-xrc1swqbm","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/intel-s-cheapest-cpus-do-what-google-s-discontinued-coral-accelerators-used-to-do-and-they-cost-les-xrc1swqbm"},"datePublished":"2026-07-26T17:30:32.201Z","dateModified":"2026-09-14T08:30:16.313Z","description":"Google's Coral AI accelerators were once the go-to hardware for running local AI inference in self-hosted NVR setups like Frigate, but Google discontinued the...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e69d3023ab73780ac47ff8b9cc937746?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e69d3023ab73780ac47ff8b9cc937746?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"XDA Developers","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"XDA Developers","logo":"https://media.daily.dev/image/upload/s--kCg6yyAP--/f_auto,q_auto/v1774964407/logos/xda-developers?_a=BAMAMiWQ0","url":"https://daily.dev/sources/xda-developers"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/intel-s-cheapest-cpus-do-what-google-s-discontinued-coral-accelerators-used-to-do-and-they-cost-les-xrc1swqbm","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":5},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"raspberry-pi,object-detection","timeRequired":"PT4M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"XDA Developers","item":"https://daily.dev/sources/xda-developers"},{"@type":"ListItem","position":3,"name":"Intel's cheapest CPUs do what Google's discontinued Coral accelerators used to do, and they cost less"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/intel-s-cheapest-cpus-do-what-google-s-discontinued-coral-accelerators-used-to-do-and-they-cost-les-xrc1swqbm#faq","mainEntity":[{"@type":"Question","name":"Why did Frigate stop recommending Google Coral TPU accelerators?","acceptedAnswer":{"@type":"Answer","text":"Frigate 0.17 stopped recommending Coral accelerators because Google discontinued support for the Coral TPU lineup and most required libraries have not been updated since 2022. The GASKET driver needed for PCI-based Coral devices is not supported on modern Linux kernels, requiring workarounds. Existing Coral setups still work, but Frigate now steers GPU-less users toward Intel OpenVINO instead. Anyone rebuilding a self-hosted NVR pipeline can track shifts like this Coral-to-OpenVINO move on daily.dev."}},{"@type":"Question","name":"What can I use instead of a Google Coral accelerator for Frigate object detection?","acceptedAnswer":{"@type":"Answer","text":"Intel OpenVINO running on an inexpensive CPU like the Intel N100, or on an Intel NPU in a mini-PC, can replace a Coral accelerator for Frigate object detection, running models such as Yolov9 or MobileNet v2. It may draw slightly more power than a dedicated Coral Edge TPU, but avoids buying discontinued or unsupported hardware. Comparing self-hosted AI accelerator options gets easier when the alternatives are tracked in one place on daily.dev."}},{"@type":"Question","name":"Can a Raspberry Pi 5 handle AI-based object detection for security cameras on its own?","acceptedAnswer":{"@type":"Answer","text":"A Raspberry Pi 5 can run AI detection tasks on its own CPU, but doing so requires lowering stream resolution and configuring large masks to keep CPU utilization from maxing out. Adding a Raspberry Pi AI Kit with a Hailo-8L accelerator offloads this inference work, allowing full-resolution camera streams without straining the SBC's processor. Developers weighing SBC hardware add-ons for local AI workloads can follow updates like this on daily.dev."}}]} ```