Run AI on your own terms! Ollama lets you run powerful language models right on your own computer or server. No cloud, no API costs, no data leaving your network. This add-on connects AI Engine to your local Ollama installation.
Perfect for privacy-conscious users, developers who want to experiment, or anyone who wants to avoid per-token API fees. It’s fast, it’s free (after setup), and your data stays with you.
Features
- Chat Models — Run Llama, Mistral, Gemma, and other language models locally.
- Vision Models — Use LLaVA and Llama 3.2 for image understanding in your chatbot.
- Embeddings — Generate embeddings locally with nomic-embed-text or all-minilm for RAG and semantic search.
- Function Calling — Use tools and function calling with supported models.
- Connection Check — Verify your Ollama connection before use.
How to use it
First, install Ollama on your computer or server. Then, in AI Engine’s Environments settings, select the Ollama type and enter your endpoint (usually http://127.0.0.1:11434 for local installations).
Click Refresh Models to see available models, then use your new Ollama environment anywhere in AI Engine!
Recommended Models
- Llama 3.2 — Great for chat, vision, and function calling
- LLaVA — Specialized vision model for image understanding
- Mistral — Fast and capable chat model
- nomic-embed-text — Embedding model for RAG
For more models and setup instructions, visit the official Ollama website.