Knowledge Base

Frequently Asked Questions

Everything you need to know about the product, billing, security, and how the Toffee AI context protocol works.

General

What is AI Context Transfer?

AI context transfer is the process of capturing the knowledge, decisions, and preferences accumulated during a conversation with one AI model and porting that information to another AI session or platform, preserving continuity without manually re-explaining everything.

Which AI platforms are supported?

Toffee currently supports ChatGPT (GPT-4o, o1), Anthropic Claude (Sonnet 3.5, Opus), Google Gemini (1.5 Pro), Microsoft Copilot, and Perplexity AI. We are actively adding support for Grok and local models like Llama 3.

Is Toffee a browser extension or an API?

Both. The easiest way to use Toffee is via our browser extension for Chrome, Edge, and Brave. For developers and enterprises, we provide an API to integrate our semantic compression engine directly into your applications.

Privacy & Security

Does Toffee see my AI conversations?

By default, no. Toffee operates on a local-first architecture. When you capture and compress a conversation using the browser extension, all processing happens locally on your device. Data is stored in your browser's IndexedDB.

How does Cloud Sync work securely?

If you opt-in to Cloud Sync to access your bundles across devices, your data is encrypted on your device using AES-256-GCM before being sent to our servers. We cannot read the contents of your synchronized context bundles.

Is Toffee compliant with enterprise security policies?

Yes. Our enterprise tier includes SSO, SAML integration, detailed audit logs, and the ability to deploy the compression engine in your own VPC or private cloud infrastructure.

Technical

How much does semantic compression reduce token usage?

On average, Toffee's semantic compression reduces conversation token usage by 60-80% while preserving over 95% of actionable context (decisions, entities, code, and preferences).

What happens to code blocks during transfer?

Toffee identifies the final, working version of any code artifact. Intermediate attempts, debugging iterations, and superseded implementations are discarded to save tokens, but the final solution and its context are preserved perfectly.

Can I self-host the Toffee API?

Yes, self-hosting is available for our Enterprise customers. Please contact our sales team at enterprise@toffee.ai for more information.

Still have questions?

Our support team is ready to help you with any technical or billing issues.

Contact Support