![\"Picture](\"https:\/\/www.newsbeep.com\/il\/wp-content\/uploads\/2026\/03\/VR_Running_AI_on_Pi_P2_Ollama_html_b7be4f5b_s.ashx.jpeg\")
<\/a>
\n [Click on image for larger view.]<\/p>\nOllama emerged in
\n response to a movement toward greater access to AI and the
\n decentralization of inference workloads. It really started to gain
\n traction in 2023 and accelerated rapidly through 2024 and 2025 as
\n local model performance improved dramatically. As quantization
\n techniques, optimized runtimes, and efficient architecture made it
\n possible to run capable models on consumer-grade hardware, Ollama
\n quickly positioned itself as an easy way for those with limited
\n resources to get started with AI. Its rapid adoption has been
\n fueled by its simplicity, active open-source community, and growing
\n ecosystem of supported models and integrations.\n<\/p>\n
Ollama Commands
Here are a few of
\n the commands I used with Ollama. These examples use llama3 as the
\n LLM, but you can use other LLMs.<\/p>\n
\n Command<\/p>\n
\n Example<\/p>\n
\n What
\n It Does<\/p>\n
\n ollama
\n pull<\/p>\n
\n ollama
\n pull llama3<\/p>\n
\n Downloads
\n a model from the Ollama registry and stores it locally for offline
\n use.<\/p>\n
\n ollama
\n run<\/p>\n
\n ollama
\n run llama3<\/p>\n
\n Launches
\n an interactive chat session with a model. It will download the
\n model if needed.<\/p>\n
\n ollama
\n list<\/p>\n
\n ollama
\n list<\/p>\n
\n Lists
\n all models currently installed on the local system.<\/p>\n
\n ollama
\n rm<\/p>\n
\n ollama
\n rm llama3<\/p>\n
\n Deletes
\n a locally stored model to free disk space.<\/p>\n
\n ollama
\n show<\/p>\n
\n ollama
\n show llama3<\/p>\n
\n Displays
\n detailed information about a model, including parameters and
\n configuration.<\/p>\n
\n ollama
\n serve<\/p>\n
\n ollama
\n serve<\/p>\n
\n Starts
\n the local Ollama API server for programmatic access and
\n integrations.<\/p>\n
\n ollama
\n ps<\/p>\n
\n ollama
\n ps<\/p>\n
\n Shows
\n currently running models and active inference processes.<\/p>\n
\n ollama
\n stop<\/p>\n
\n ollama
\n stop llama3<\/p>\n
\n Stops
\n a currently running model session.<\/p>\n
\n ollama
\n create<\/p>\n
\n ollama
\n create my-model -f Modelfile<\/p>\n
\n Builds
\n a custom model using a Modelfile configuration.<\/p>\n
\n ollama
\n cp<\/p>\n
\n ollama
\n cp llama3 my-model<\/p>\n
\n Copies
\n an existing model locally, often used as a base for customization.<\/p>\n
\nTom’s Tip: Use \/bye to exit ollama.<\/p>\n
\n Installing
\n a Local LLM on Raspberry Pi
Installing and
\n running an LLM on Raspberry Pi is relatively simple using Ollama.\n<\/p>\n
Deciding which
\n LLM to run is more difficult. The “religious” wars over LLMs
\n are intense and make past IT wars, such as Windows vs. Mac or
\n file vs. block storage, seem minor compared to the opinions about LLM
\n selection. After far too much research, I decided to start with
\n qwen2.5.<\/p>\n
Qwen is an
\n interesting LLM and made quite a ruckus when it was first released,
\n as it is an LLM from the Chinese company Alibaba. The Qwen family of
\n LLMs were, designed to deliver strong performance across reasoning,
\n coding, multilingual understanding, and general-purpose text
\n generation.\n<\/p>\n
<\/p>\n","protected":false},"excerpt":{"rendered":"How-To Running AI on a Raspberry Pi, Part 2: Running AI on a Pi in Under 5 minutes…\n","protected":false},"author":2,"featured_media":366402,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[6],"tags":[75429,5130,20721,2934,15083,176553,176559,85,46,18529,134,176556,176555,140,176557,35362,125,21652,176558,58642,176554],"class_list":{"0":"post-366401","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-technology","8":"tag-backup","9":"tag-cloud","10":"tag-containers","11":"tag-edge","12":"tag-hybrid-cloud","13":"tag-hyper-v","14":"tag-hyperconverged","15":"tag-il","16":"tag-israel","17":"tag-kubernetes","18":"tag-microsoft","19":"tag-sd-wan","20":"tag-sdn","21":"tag-security","22":"tag-serverless","23":"tag-storage","24":"tag-technology","25":"tag-veeam","26":"tag-virtual-desktops","27":"tag-vmware","28":"tag-vsphere"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/posts\/366401","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/comments?post=366401"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/posts\/366401\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/media\/366402"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/media?parent=366401"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/categories?post=366401"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/il\/wp-json\/wp\/v2\/tags?post=366401"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
![\"Picture](\"https:\/\/www.newsbeep.com\/il\/wp-content\/uploads\/2026\/03\/VR_Running_AI_on_Pi_P2_Ollama_html_23b467bb_s.ashx.jpeg\")
<\/a>