Artificial intelligence is now available to everyone, and not just on powerful servers. With Ollama<\/a><\/strong>you can run language models (LLMs) directly in local<\/strong>. Before installing, update your Pi :<\/p>\n\n\n Then restart :<\/p>\n\n\n Ollama offers an official installation script. Download it and run it:<\/p>\n\n\n \u26a1 This installs Ollama and automatically configures the system service.<\/p>\n\n\n\n Once installed, check that Ollama is active:<\/p>\n\n\n You should see a version number displayed.<\/p>\n\n\n\n You can also check that the service is running:<\/p>\n\n\n To execute LLaMA 2<\/strong>type :<\/p>\n\n\n The template will be automatically downloaded (several hundred MB in size).<\/p>\n\n\n\n Example with a lighter model (TinyLlama)<\/p>\n\n\n \ud83d\udc49 On the Raspberry Pi, prefer the streamlined and quantified models<\/strong> (Q4, Q5), otherwise performance will be very slow.<\/p>\n\n\n\n Once the model has been launched, you can chat directly in the terminal.<\/p>\n\n\n\n Example:<\/p>\n\n\n To exit :<\/p>\n\n\n Ollama integrates a Local API<\/strong>. You can access it on the port Example with \ud83d\udc49 Result: a JSON response containing the generated text.<\/p>\n\n\n\n This allows you to integrate Ollama into your applications, websites or home automation projects<\/strong>.<\/p>\n\n\n\n
This tutorial shows you how to install Ollama on a Raspberry Pi 5 (or Pi 4)<\/strong>and how to start using your first AI models locally.<\/p>\n\n\n\n1. Prerequisites<\/h2>\n\n\n\n
Recommended equipment<\/h3>\n\n\n\n
\n
Software<\/h3>\n\n\n\n
\n
2. System update<\/h2>\n\n\n\n
\nsudo apt update && sudo apt upgrade -y\n<\/pre><\/div>\n\n\n
\nsudo reboot\n<\/pre><\/div>\n\n\n
3. Installing Ollama on Raspberry Pi<\/h2>\n\n\n\n
\ncurl -fsSL https:\/\/ollama.com\/install.sh | sh\n<\/pre><\/div>\n\n\n
4. Check installation<\/h2>\n\n\n\n
\nollama --version\n<\/pre><\/div>\n\n\n
\nsystemctl status ollama\n<\/pre><\/div>\n\n\n
5. Download and run an IA model<\/h2>\n\n\n\n
Example with LLaMA 2<\/h3>\n\n\n\n
\nollama run llama2\n<\/pre><\/div>\n\n\n
Example with Mistral<\/h3>\n\n\n
\nollama run mistral\n<\/pre><\/div>\n\n\n
\nollama run tinyllama\n<\/pre><\/div>\n\n\n
6. Interact with Ollama<\/h2>\n\n\n\n
\n> Hello, who are you?\nI'm a language model powered by Ollama!\n<\/pre><\/div>\n\n\n
\n\/bye\n<\/pre><\/div>\n\n\n
7. Using Ollama via API REST<\/h2>\n\n\n\n
11434<\/code>.<\/p>\n\n\n\ncurl<\/code> :<\/p>\n\n\n\ncurl http:\/\/localhost:11434\/api\/generate -d '{\n \"model\": \"llama2\",\n \"prompt\": \"Explain the Raspberry Pi in 3 sentences.\"\n}'\n<\/pre><\/div>\n\n\n8. Optimizations for Raspberry Pi<\/h2>\n\n\n\n