Tresor is built for the documents, conversations, and decisions that have to stay private.
Real capabilities. The same guarantee on every one.
Upload it once. Ask it anything.
Files are parsed, chunked, and embedded inside the enclave, and every answer comes back with citations to the exact passage and a shield proving which model produced it. PDF, DOCX, XLSX, CSV, PPTX, RTF, ODF, HTML, images.
Browse the web without the web seeing you.
Search runs from inside the enclave with your approval on each query, and the stripped request goes to DuckDuckGo without your identity ever attached. Edit your query before sending it.
Collaborate on the documents you'd never email.
Each member encrypts and decrypts on their own device, so workspace keys are unwrapped locally and Tresor's servers only ever relay encrypted blobs. Roles: Owner, Admin, Member.
Watch it reason. Then verify what happened.
Thinking mode plans, calls tools, and reflects on its own evidence step by step, and the full trace is bound to the same verification receipt as the answer. You can pick the mode and model per message.
Proof, not promises, attached to every answer.
Every response carries a signed receipt that proves it was generated inside the expected enclave, and you can verify it yourself against our public signing keys. Verify the JWS/ES256 with any JWT library.
For the chats and files that belong together.
Projects bundle conversations, files, and custom instructions under their own encryption key, so you can share a project without exposing the rest of your workspace. Share with Can View or Can Edit permissions.
Many frontier models. One privacy guarantee.
Every model runs inside a confidential enclave, so switching between Kimi K2.5, GPT-OSS, DeepSeek, Llama, Mistral, or GLM never changes how your data is protected or verified. Switching models never changes the privacy guarantee.
Public AI tools read everything you send. On-prem is private but impractical. Tresor gives you both: cloud convenience, infrastructure-grade privacy.
I can only use AI in coaching when it fully protects the sacred trust between practitioner and client; confidentiality is non-negotiable.
Tove Thyes
Transformational Coach and Energy Medicine Practitioner
As a software strategist, I treat privacy as the blueprint that lets my team turn client visions into AI people can trust.
Igor Miazek
CEO & Founder, Techs
At Dance, we build experiences people trust. Tresor’s approach to privacy-first AI is a simply a great match.
Christian Springub
CEO & Co-Founder, Dance
As a therapist, I can only use AI that protects client privacy. Tresor delivers exactly that.
Magali Cahen
Psychologist and Therapist, Independent
The only way to earn trust in AI is to make privacy a design principle, not a feature. Tresor’s system does exactly that.
Ingmar Schuster
CEO & Co-Founder, Provolut
In the AI era, rigorous risk-based security and GDPR-aligned privacy are non-negotiable foundations for any trustworthy system.
Davy Cox
Founder, Brainframe
For individuals getting started with private AI.
0€
Advanced features for individuals and freelancers.
25€
/month
Collaboration for secure teams with shared control.
39€
/seat/month
Custom features with enterprise-grade compliance and support.
Custom
The best of open source, isolated and verifiable in Zero-Access TEEs.
Executive Summary
Tresor is built on a simple promise: your conversations belong to you, not us. Every message you type is protected by end-to-end encryption and processed only inside secure computing environments that even Tresor cannot inspect. Teams can now collaborate inside shared workspaces without ever handing Tresor access to their plaintext. This whitepaper explains the principles and safeguards behind Tresor’s zero-access design, showing how we deliver practical confidentiality without trade-offs in usability.
