Privacy in Venice

Every Venice user gets anonymous and private AI by default. Pro users unlock hardware-verified and end-to-end encrypted inference for the strongest possible protection.

Privacy Architecture

The Venice proxy is at the core of every privacy mode

Every AI interaction on Venice passes through the same proxy architecture. Your conversation history stays on your device. Requests are relayed through the proxy without being stored. The four privacy modes build on top of this foundation, adding progressively stronger protections at the GPU layer.

Venice Proxy

Four Privacy Modes

Every model on Venice falls into one of four privacy categories. Each comes with different protections and trade-offs.

Anonymous

Venice proxies your request to leading frontier model providers. Your identity is obscured from the provider, but you should assume the provider is storing your content.

  • Access to leading frontier models (GPT, Claude, Gemini, Grok)
    − Provider sees and likely saves your prompt

Private

Default

Inference runs on Venice-controlled GPUs or zero-data-retention partner infrastructure. Privacy is enforced by contractual commitments — no prompt or response is stored.

  • No data retention, wide open-source model selection
    − Trust-based — relies on Venice and partners honoring contracts

TEE

Pro

Trusted Execution Environment. Inference runs inside a hardware-isolated enclave operated by external TEE partners (NEAR AI Cloud and Phala Network). GPU providers cannot access your prompts. Verified by remote attestation.

  • Hardware-verified privacy, GPU operator cannot see prompts
    − Responses may be slower; fewer models available

E2EE

Pro

End-to-End Encrypted. Your prompt is encrypted on your device, remains encrypted through Venice, and is only decrypted inside a verified TEE. Verifiably and entirely private from all parties.

  • Strongest privacy — no one (including Venice) can see your data
    − No web search or memory; responses may be slower

Choose your privacy for every conversation

Every AI model on Venice falls into one of four privacy modes. Switch between them based on what matters most — model selection, speed, or verifiable encryption.

Select a privacy mode to see how it protects you

  • Anonymous: Identity obscured from provider
  • Private: Zero data retention, contract-enforced
  • TEE: Hardware-isolated GPU processing
  • E2EE: Encrypted end-to-end, fully verifiable

Your DeviceVerified

Local storage only

Your conversation history stays on your device. Venice never stores it on its servers.

Secure relay via Venice

Your request passes through Venice's proxy over standard SSL/TLS and is relayed without being stored.

Provider policies apply

The model provider processes your prompt under its own retention and privacy policies. Venice does not control this layer.

Venice Privacy FAQ

Answers about Venice's privacy philosophy, proxy architecture, trust assumptions, data collection, TEE, E2EE, attestation, model availability, and providers.

Philosophy

What is Venice's privacy philosophy?

Architecture

How does the Venice proxy work?

Trust

What are the trust assumptions in proxy-based modes?

Transparency

What does Venice store?

Mode

What is TEE?

What is E2EE?

Models

Which models are TEE and E2EE?

Verification

What is attestation and how does it work?

Product

What features are limited in E2EE mode?

Providers

Who are the TEE & E2EE model providers?

TEE and E2EE are Pro features

Anonymous and Private modes are available to all users. TEE and E2EE models require a Venice Pro subscription. Upgrade to access hardware-verified and end-to-end encrypted inference.