Is data encrypted before leaving my machine?

Yes. The Privatemode proxy runs locally and encrypts all data before it leaves your machine. It also performs remote attestation to verify the integrity of the Privatemode service, ensuring end-to-end protection for every request.

Does Privatemode store or retain my conversations?

No. Privatemode is designed to never learn from your data. All inference requests are processed confidentially using confidential computing, and no prompts or responses are retained after processing.

Can I still use local models alongside Privatemode?

Yes. GPT4All lets you switch between local and remote models freely. You can use Privatemode for tasks that benefit from frontier cloud models and local models for quick, offline work.

How does this differ from using standard cloud providers like OpenAI?

Standard cloud providers process your prompts in plaintext on their servers. Privatemode uses confidential computing hardware so your data is encrypted even during inference. The AI provider and Privatemode itself never see your data in the clear.

Integrating Privatemode AI with GPT4All

GPT4All is built around a simple promise: your conversations stay on your device. Privatemode extends that promise to the cloud, letting you access frontier models without sending your data to any provider.

Introduction

Integrating GPT4All

Benefits

Why integrating GPT4All

Introduction

Bring frontier AI to GPT4All without sacrificing privacy

Local models or cloud-grade capability

GPT4All users choose local AI because they don't want data leaving their device. But local models are constrained by your hardware's memory and compute. Privatemode lets you access state-of-the-art cloud models while maintaining the same data control GPT4All users expect.

Zero data exposure to the provider

Every prompt and response is encrypted end-to-end before leaving your machine. The Privatemode proxy handles encryption and remote attestation locally, so the AI provider never sees your conversations in plaintext.

No changes to your GPT4All workflow

GPT4All supports remote model providers natively. Add Privatemode as a custom OpenAI-compatible endpoint in settings, and your familiar chat interface works exactly the same, now powered by frontier cloud models with full encryption.

Benefits

Why use Privatemode AI with GPT4All?

GPT4All interface with end-to-end encrypted cloud AI

Integrating Privatemode into GPT4All brings cloud-AI capabilities to your desktop chat while keeping your conversations confidential. Every message is encrypted through end-to-end encryption and confidential computing. Privatemode never learns from your data.

Access frontier models beyond local hardware limits

Local models are capped by your device's memory and GPU. With Privatemode, you can access state-of-the-art LLMs that far exceed what local hardware can run, while keeping the same privacy guarantees.

Set up in minutes, no infrastructure required

Run the Privatemode proxy via Docker, add it as a remote provider in GPT4All's settings, and start chatting. Your local models stay available alongside Privatemode for offline use.

How to get started

How to set up Privatemode in GPT4All

Get your API key

If you don't have a Privatemode API key yet, you can generate one for free here.

docker run -p 8080:8080 ghcr.io/edgelesssys/privatemode/privatemode-proxy:latest --apiKey <your-api-key>

Run the Privatemode proxy

The proxy verifies the integrity of the Privatemode service using confidential computing-based remote attestation. The proxy also encrypts all data before sending and decrypts data it receives.

Set up GPT4All

Download GPT4All from the official website: https://www.nomic.ai/gpt4all and open the application on your computer.

Integrate Privatemode with GPT4All

FAQ

Frequently asked questions about using Privatemode with GPT4All

Minimal changes. GPT4All supports remote OpenAI-compatible providers. You add Privatemode as a remote model endpoint in the settings, and the chat interface works as usual. Your local models remain available for offline use alongside Privatemode.