Configure Providers

Providers connect Scope to LLM services like OpenAI, Anthropic, Google, and AWS Bedrock. This guide covers adding providers, testing connections, managing models, and updating configurations.

Supported Providers

Scope supports the following LLM providers:

Provider	Description
OpenAI	GPT-4o, GPT-4, GPT-3.5 Turbo, and other OpenAI models
Anthropic	Claude 3 Opus, Sonnet, Haiku, and other Anthropic models
Google	Gemini Pro, Gemini Ultra, and other Google AI models
AWS Bedrock	Access to multiple model providers through AWS Bedrock

Add a Provider

1. Go to Settings > Providers
2. Click Add Provider
3. Select the provider from the available list
4. Configure the provider:
   • Display name — a human-readable label (e.g., "OpenAI Production")
   • API key — your provider's API key
   • Config name (optional) — defaults to default. Use a custom name to create multiple configurations of the same provider (e.g., production, staging)
   • Base URL (optional) — custom API endpoint for proxies or self-hosted deployments
5. Click Test Connection to verify the API key
6. Save the provider

warning

Provider API keys are stored encrypted. Only users with provider management permissions can view or update provider configurations.

Test a Connection

Before saving, always test the connection to verify your API key is valid:

1. Fill in the provider name and API key
2. Click Test Connection
3. Scope makes a lightweight validation request to the provider
4. A success or failure message appears

Enable Models

After adding a provider, enable the specific models you want to use:

1. Open the provider from the provider list
2. Click Available Models to see the model catalog
3. Enable the models you need (e.g., gpt-4o, gpt-3.5-turbo)
4. Optionally set a custom display name for each model

Enabled models appear in the test panel's model selector and can be used for prompt execution.

Model Properties

Each model has the following properties:

Property	Description
external_model_id	The provider's model identifier (e.g., gpt-4o)
display_name	Your custom label for the model
enabled	Whether the model is available for use
capabilities	Feature flags: function calling, vision, streaming, JSON mode
context_window	Maximum input token count
max_output_tokens	Maximum output token count
input_cost_per_1m	Cost per million input tokens
output_cost_per_1m	Cost per million output tokens

Update a Provider

To modify a provider's settings:

1. Open the provider from Settings > Providers
2. Update the fields:
   • Display name — change the label
   • Enabled — toggle the provider on or off
   • Base URL — update the custom endpoint
3. Save changes

info

To update the API key, you'll need to delete the existing provider and create a new one, or use the API to update it directly.

Disable or Delete a Provider

• Disable — toggle the provider off in settings. The configuration is preserved but the provider won't appear in the test panel's provider selector
• Delete — permanently remove the provider and all its model configurations

Multiple Configurations

You can configure the same provider type multiple times using different config_name values. This is useful for:

• Separating production and staging API keys
• Using different base URLs for proxy configurations
• Managing separate rate limits or billing accounts