← Back to Home

Privacy Policy

Last Updated: January 20, 2026

1. Data We Collect

  • Account Information: We use Clerk for authentication. We store your email and user ID.
  • Project Data: Descriptions, requirements, and configurations you input into the dashboard.
  • API Keys: If you provide your own API keys, they are stored using industry-standard AES-256-GCM encryption. We cannot view your raw keys.
  • Prompts: The text prompts you submit for testing.

2. How We Use Your Data

We use your data solely to provide the LLM benchmarking service.We do not use your prompts or data to train our own models.

3. Data Sharing & Sub-processors

To perform benchmarks, YSelector must transmit your prompts to the third-party Model Providers you select. By using the service, you authorize data transmission to:

  • OpenAI (via Azure or Direct)
  • Anthropic (via Azure or Direct)
  • Google (Vertex AI / Gemini)
  • Meta (via Azure MaaS)

We also use Supabase for database hosting and Inngest for background job processing.

4. Data Security

We employ strict security measures. API keys are encrypted at rest. All data in transit is protected via TLS 1.3. Our application filters output for basic safety, but we rely on the safety filters of the underlying model providers.

5. User Rights

You may delete your projects and test history at any time via the Dashboard. Upon account deletion, all your data, including encrypted API keys, is permanently removed from our database.