AI Configuration Profiles

Sypha allows you to manage multiple Configuration Profiles, enabling rapid transitions between different AI infrastructures. These profiles can be customized per operational mode, ensuring you always have the right model for the right task.

[!TIP] Utilize profiles to partition your work by provider (e.g., Anthropic vs. OpenAI) or cost requirements (e.g., usage of free vs. high-performance premium models) without needing to re-configure keys manually.

How Profiles Function

Every configuration profile encapsulates:

• Provider Settings: Seamlessly manage OpenAI, Anthropic, OpenRouter, Google, and more.
• Secure Credentials: Encrypted management of your API keys.
• Model Selection: Designation of specific engines, such as Claude 3.7 Sonnet or GPT-4o.
• Logic Parameters: Fine-tune Temperature and computational reasoning budgets.

Profile Administration

1. Creation: Navigate to Settings (gear icon) -> Providers and select the "+" icon to initialize a new set.
2. switching: Use the profile selector found in either the Settings dashboard or at the top of the Chat Interface.
3. Prioritization (Pinning): Hover over a profile in the list and click the pin icon to ensure your most-used settings remain at the top.
4. Modification/Removal: Use the edit icon for renaming or the delete icon to remove obsolete profiles (a minimum of one profile must always exist).

Mode Association

Within the Prompts configuration, you can link specific profiles to specific modes. For example, you might prefer Architect mode to use high-reasoning models while Code mode uses faster, cost-efficient alternatives. Sypha also intuitively recalls the most recent profile used for each mode.

Security Standards

All API credentials are encrypted and stored within VS Code's native Secret Storage, ensuring they are never exposed in plaintext within your project files.