Notely

AI Setup ​

Configure models and API settings in the AI → AI Settings menu.

1. Provider Setup ​

Configure separate models for text generation and semantic features:

  • Text Provider: Generates chat responses and executes text rewrites. Supports Local API Endpoints (e.g. Ollama, LM Studio) and OpenAI.
  • Embeddings Provider: Generates vector representations of your notes for meaning-based search. Supported via HuggingFace token configurations.
  • Connection Test: Click Test next to either service to verify authentication and connection parameters.

2. Feature Toggles ​

  • Learn User Patterns: Allows the system to capture local usage context to improve suggestions.
  • Generate Embeddings: Builds a local vector database of note contents for search.
  • Discover Relationships: Analyzes references to map semantic proximity in the workspace graph.

3. Advanced Parameters ​

Fine-tune AI responses:

  • Temperature: Controls creativity. Lower values yield structured, predictable output; higher values yield creative variations.
  • Max Tokens: Bounds response length to manage speed and resource consumption.
  • Local Storage Path: Shows where cache data is saved, with a button to Clear Learned Data.