Models

Models in Rhesis are used for two primary purposes:

• Test Generation: Automatically generate test cases based on your requirements (see Tests)
• Evaluation (LLM as Judge): Evaluate AI responses using metrics powered by LLMs (see Metrics)

You can configure default models for each purpose in the model settings, or select specific models when creating individual metrics or test cases.

Default Model

Rhesis provides a default managed model that requires no configuration:

• Rhesis Default: Default Rhesis-hosted model
  • No API key required
  • Pre-configured for both generation and evaluation
  • Ready to use immediately

Supported Providers

Connecting a Model Provider

1. Click “Add Model” on the Models page
2. Select a provider (e.g., Google Gemini, OpenAI, Anthropic)
3. Fill in the required fields:
   • Connection Name: Unique identifier for this connection
   • Model Name: Specific model to use (e.g., “gemini-1.5-pro”, “gpt-4-turbo”)
   • API Key: Your API key from the provider’s dashboard
4. (Optional) Configure additional settings:
   • Custom Headers: Add HTTP headers required for your API calls
   • Default for Test Generation: Use this model when generating test cases
   • Default for Evaluation: Use this model for metrics evaluation
5. Click “Test Connection” to verify your configuration
6. Click “Save” to add the model

Provider-Specific Connection Fields

Some providers require endpoint configuration in addition to model name:

Polyphemus Access Control (v0.6.6+)

Polyphemus uses an explicit request-and-review workflow. If access is not granted yet, the model card shows Access Required and a Request Access action.

Request Workflow

1. Open Models and click Request Access on the Polyphemus card.
2. Submit request details in the access modal.
3. The request is recorded and review notification is sent.
4. After approval, your account is marked as verified and the model becomes available.

Request Fields