AI & Ollama Integration

Pinnacle leverages local Artificial Intelligence to provide deep document insights without compromising your privacy.

Why Local AI?

Most document tools send your PDFs to cloud-based LLMs (like GPT-4). Pinnacle takes a different path:

Private by Default: Your data never leaves your machine.
Cost Effective: No subscription fees or API costs.
Offline Capable: Summarize and chat with your docs even without internet.

Setting Up Ollama

Pinnacle uses Ollama as the primary backend for local intelligence.

Download & Install: Visit ollama.com to download and install Ollama for your OS.
Pull the Model: Open your terminal and pull the recommended model for Pinnacle:
bash
```
ollama run llama3
```
Note: You can also use other models like mistral or phi3 depending on your machine's resources.
Pinnacle Connection: Once Ollama is running, Pinnacle will automatically detect it on localhost:11434.

Features Powered by AI

Insight Pane: Automatically generates summaries, extracts key dates, and identifies organizations from your PDFs.
Contextual Chat: Ask questions about your currently open document.
Document Naming: Uses AI to suggest descriptive titles based on content analysis.
Knowledge Graph: Powers the entity extraction engine that builds relationships between your documents.