Claude Code CLI

Claude Code is Anthropic’s official agentic coding tool that runs in your terminal. This guide shows how to configure Claude Code to use LLM Gateway.

Prerequisites

  • Claude Code installed (npm install -g @anthropic-ai/claude-code)
  • LLMGW API token (see User Tokens)

Configuration

Claude Code connects to LLMGW via the AWS Bedrock endpoint. Add the following environment variables to your shell configuration file (.bashrc, .zshrc, or similar):

# LLMGW Configuration for Claude Code
export ANTHROPIC_DEFAULT_HAIKU_MODEL="anthropic.claude-haiku-4-5-20251001-v1:0-native"
export ANTHROPIC_DEFAULT_SONNET_MODEL="anthropic.claude-haiku-4-5-20251001-v1:0-native"
export ANTHROPIC_BEDROCK_BASE_URL="https://<llmgw-deployment-url>/aws-bedrock"
export ANTHROPIC_AUTH_TOKEN="your-llmgw-token-here"
export CLAUDE_CODE_USE_BEDROCK=1
export CLAUDE_CODE_SKIP_BEDROCK_AUTH=1

After adding these variables, reload your shell:

source ~/.zshrc  # or ~/.bashrc

Environment Variables Explained

Variable Description
ANTHROPIC_DEFAULT_HAIKU_MODEL Model ID for fast/cheap operations
ANTHROPIC_DEFAULT_SONNET_MODEL Model ID for complex tasks
ANTHROPIC_BEDROCK_BASE_URL LLMGW AWS Bedrock endpoint
ANTHROPIC_AUTH_TOKEN Your LLMGW API token
CLAUDE_CODE_USE_BEDROCK Enable Bedrock mode
CLAUDE_CODE_SKIP_BEDROCK_AUTH Skip AWS auth (LLMGW handles authentication)

Model Selection

Claude Code uses two models:

  • Haiku model - Used for fast, simple operations (file summaries, quick questions)
  • Sonnet model - Used for complex coding tasks (code generation, refactoring)

Cost Saving Tip: For testing purposes, set both models to Haiku to minimize costs. You can upgrade to Sonnet later when you need better quality for complex tasks.

Available Models

Check the Available Models page for the current list. Common options:

Model Use in Variable
Claude Haiku 4.5 (Recommended for testing) anthropic.claude-haiku-4-5-20251001-v1:0-native
Claude Sonnet 4.5 (For production/complex tasks) anthropic.claude-sonnet-4-5-20250929-v1:0-native

Verification

To verify your configuration is working:

  1. Start Claude Code:

    claude

  2. Ask a simple question to test connectivity:

    What is 2 + 2?

  3. Check the LLMGW Admin Portal to confirm requests are being logged.

Troubleshooting

“Invalid API Key” Error

If you see authentication errors:

  1. Verify your token hasn’t expired
  2. Regenerate a new token if needed from the Admin Portal
  3. Ensure you’ve copied the full token string
  4. Verify the endpoint URL is correct

“Model not found” Error

If you get “model not found” errors:

  1. Check the Available Models page for correct model IDs
  2. Ensure model IDs match exactly (including the -native suffix)

Connection Timeout

If you experience connection timeouts:

  1. Check network connectivity to LLMGW
  2. If using a VPN, ensure it allows access to LLMGW endpoints

Slow Responses

If responses are slower than expected:

  1. Consider using a faster model for simple tasks
  2. Check if the model is under high load (contact administrator)
  3. Try a different model from the same group

Spend Limit Exceeded

If you see an error like:

API Error: 403 Spend limit exceeded for project_user name: <email> · Please run /login

This means you’ve reached your daily/weekly/monthly spend limit. To resolve:

  1. Wait for the limit to reset (check with your administrator for reset period)
  2. Contact your administrator/PM to increase limits if needed

Best Practices

  1. Use environment variables for sensitive configuration like API tokens
  2. Verify connectivity - Check the Admin Portal to confirm requests are being logged
  3. Choose appropriate models - Haiku for simple tasks saves costs
  4. Reload shell after changing environment variables