Anthropic

The Anthropic API provider enables direct integration with Claude 3.7 Sonnet and other Anthropic models. This guide covers setup, configuration, authentication, and usage options for connecting to Anthropic’s API.

Overview

The Anthropic API provider allows direct access to Anthropic’s Claude models, including Claude 3.7 Sonnet. This provider offers a streamlined integration with robust authentication, flexible configuration options, and support for advanced features like thinking mode.

Authentication

The provider uses API key authentication to access Anthropic’s services. You’ll need to obtain an API key from Anthropic to use this provider:

• Environment Variable: Set ANTHROPIC_API_KEY to provide your credentials.
• Command Line: Pass your API key using the appropriate flag.

For more details on obtaining and managing Anthropic API keys, refer to Anthropic’s documentation:

• Anthropic API Documentation

Example Usage

export ANTHROPIC_API_KEY='yourkey'

Below is a basic example of how to use the provider with a connection configuration file:

./gateway discover \
  --ai-provider anthropic \
  --config connection.yaml

Endpoint Configuration

By default, the provider connects to Anthropic’s standard API endpoint. You can specify a custom endpoint using one of the following methods:

1. Configuration File: Include the endpoint in your connection.yaml
2. Environment Variable: Set the ANTHROPIC_ENDPOINT variable.

Example with a custom endpoint:

export ANTHROPIC_ENDPOINT="https://custom-anthropic-endpoint.example.com"
./gateway discover \
  --ai-provider anthropic \
  --config connection.yaml

Model Selection

By default, the Anthropic provider uses claude-3-7-sonnet-20250219. You can specify a different model using one of the following methods:

1. Command-line Flag: Use the --ai-model flag.
2. Environment Variable: Set the ANTHROPIC_MODEL_ID.

Examples:

# Specify model via command line
./gateway discover \
  --ai-provider anthropic \
  --ai-model claude-3-7-sonnet-20250219 \
  --config connection.yaml

# Or via environment variable
export ANTHROPIC_MODEL_ID=claude-3-7-sonnet-20250219
./gateway discover \
  --ai-provider anthropic \
  --config connection.yaml

Advanced Configuration

Reasoning Mode

Enable Claude’s “thinking” mode for complex reasoning tasks:

./gateway discover \
  --ai-provider anthropic \
  --ai-reasoning=true \
  --config connection.yaml

When reasoning mode is enabled, the provider:

• Activates Claude’s thinking capability
• Allocates a thinking token budget of 4096 tokens
• Sets the temperature to 1.0

Response Length Control

Control the maximum token count in responses:

./gateway discover \
  --ai-provider anthropic \
  --ai-max-tokens 8192 \
  --config connection.yaml

If not specified, the default maximum token count is 64000 tokens.

Temperature Adjustment

Adjust the randomness of responses with the temperature parameter:

./gateway discover \
  --ai-provider anthropic \
  --ai-temperature 0.5 \
  --config connection.yaml

Lower values produce more deterministic outputs, while higher values increase creativity and randomness.

Usage Costs

The Anthropic provider includes cost estimation for Claude Sonnet models:

• Input tokens: $3.75 per 1 million tokens
• Output tokens: $15.00 per 1 million tokens

Costs are calculated based on the actual token usage for each request.

Recommended Best Practices

• API Key Security: Use environment variables for sensitive settings like API keys.
• Manage Token Count: Set a reasonable maximum token count to control costs.
• Enable Reasoning Mode: Use for complex analytical tasks.

Additional Resources

• Anthropic API Documentation