Settings

— AI provider & intelligence config

Enable AI Features

When disabled, all LLM enrichment actions are skipped. Extractive summaries still run on ingest. Provider settings can be configured but remain inactive.

AI Provider

Provider

Model

Examples: llama3, mistral, phi3, gemma2

Embedding Model(OpenAI — used for semantic search & related events)

Recommended: text-embedding-3-small (1536 dims). Requires OPENAI_API_KEY.

Temperature — 0.20

0 — Deterministic1 — Creative

ℹOllama must be running locally at http://localhost:11434 with the selected model pulled.

Embeddings

Generate embeddings for all events that are missing them (up to 20 at a time). Requires AI enabled and OPENAI_API_KEY.