Complete Privacy

Local LLMs + Pontius

Your AI, your hardware, your rules. Run AI-powered email on your own hardware. No cloud APIs, no data leaving your machine.

Why local?

Your emails never leave your machine

Cloud AI is powerful, but your email content passes through external servers. With local models, everything stays on hardware you control.

Zero API calls

No external servers. No third-party access. Your email content never leaves your computer.

Offline capable

Works without internet after initial setup. Perfect for travel, air-gapped environments, or just peace of mind.

Fixed costs

No per-token billing, no API limits, no surprise charges. Just electricity and hardware you already own.

“Your correspondence belongs to you. Not OpenAI. Not Google. Not anyone.”

For lawyers, healthcare, finance, or anyone who takes privacy seriously — local AI means your email content stays private. No exceptions.

What you get

Full email management, fully local

Every Pontius feature works with local models. Same workflow as cloud AI, without any of the privacy tradeoffs.

Privacy-sensitive industries

Legal, healthcare, finance — industries where email content can't touch external APIs. Local models keep everything in-house.

No vendor lock-in

Switch models anytime. Today it's Llama, tomorrow it's the next open-source breakthrough. Your workflow stays the same.

Customization

Fine-tune models on your own patterns. Create a model that truly understands your communication style.

Knowledge base

Build a persistent knowledge base that works across any model. Switch from local to cloud and back — everything comes with you.

Supported tools

Works with your local setup

Ollama

Easiest way to run local models. One command to install and run.

ollama run llama3

LM Studio

Desktop app with clean UI. Great for model discovery.

GUI + local server

llama.cpp

Raw performance for power users. Maximum efficiency.

./main -m model.gguf

Any CLI tool

If it runs terminal commands, it works with Pontius.

pontius list

Getting started

Quick Start with Ollama

1. Install Pontius

$ curl -fsSL https://getpontius.com/install.sh | sh

2. Add your email account

$ pontius
# Follow the TUI prompts to add your email

3. Activate your license

$ pontius activate YOUR-LICENSE-KEY
License activated successfully.

4. Set up your local model

# Install Ollama
$ curl -fsSL https://ollama.com/install.sh | sh

# Pull a capable model
$ ollama pull llama3:8b

5. Provide context to your model

You have access to Pontius for email management. Commands:
- pontius list              # List recent emails
- pontius read <uid>        # Read an email
- pontius reply <uid> "msg" # Reply to an email
- pontius archive <uids>    # Archive emails

Execute commands to help manage the user's inbox.

Model guide

Recommended models for email

Model	Size	Best For	RAM
Llama 3 70B	40GB	Best quality	48GB+
Llama 3 8B	4.7GB	Fast, everyday use	8GB+
Mistral 7B	4.1GB	Efficient, good quality	8GB+
Mixtral 8x7B	26GB	Balanced	32GB+

Start with an 8B model and upgrade if you need better quality.

The secret sauce

It gets smarter every session

Unlike AI tools that start fresh every time, Pontius builds a persistent knowledge base. Contacts learn preferred tones. Common scenarios get templates. Rules automate the routine.

Your knowledge base is portable. Start local, switch to cloud, come back to local — everything you've built follows you.

Learn about the Knowledge Base →

Ready to get ahead?

Private AI email management on your own hardware. No compromises.

We believe you should own your software. No monthly rent extraction. No "stop paying, lose access." Buy it, own it, keep it forever.

Get Pontius — $99 Full Documentation