How INTELLI Works
INTELLI is built from the ground up as a First-Privacy AI API. Instead of relying on third-party cloud models, INTELLI ships with its own AI engine and proprietary model, optimized for local-first performance.
1. The INTELLI Engine
- Optimized for CPU — no GPU required.
- Runs anywhere — desktop, server, or edge devices.
- Privacy by design — all processing happens locally on your machine.
2. The INTELLI Model
Unlike most AI APIs that require huge downloads or external connections, INTELLI includes a compact model designed for efficiency:
- Under 3 GB total size — fits on virtually any system.
- Memory-friendly — runs with as little as 4 GB RAM.
- Offline ready — works without internet access.
This balance allows developers to get fast, reliable responses without the heavy system requirements of traditional AI frameworks.
3. Local API Architecture
- INTELLI runs as a local API service (via npm start).
- You connect using simple HTTP requests (with Bearer authentication).
- Data never leaves your environment — you are in full control.
4. Why This Matters
By designing our own engine and model, we’ve compromised nothing on privacy and efficiency. INTELLI is:
- Lightweight — works on laptops, servers, or air-gapped setups.
- Cost-efficient — no hidden API bills or GPU dependency.
- Scalable — integrate into automations, workflows, or larger systems.
Next Steps
- Test in Postman to see it in action.
- Build your first agent with n8n.