How to Configure Claude Code to Use Any Model via LLM Gateway

Claude Code works great with Claude—but what if you want to use GPT-5, Gemini, or a cheaper model for simple tasks? With LLM Gateway, you can point Claude Code at any model in our catalog. Three environment variables. No code changes.

Why Use LLM Gateway with Claude Code?

• Use any model — GPT-5, Gemini, Llama, or 180+ others with tool calling support
• Cut costs — Use GPT-4o Mini for routine tasks, Claude Opus for complex reasoning
• Track everything — See exactly what each coding session costs in your dashboard
• One config — Same setup works for all models, no provider-specific changes

Quick Setup

Set three environment variables and you're ready to go:

1export ANTHROPIC_BASE_URL=https://api.llmgateway.io
2export ANTHROPIC_AUTH_TOKEN=llmgtwy_.... # your llmgateway.io api key here
3# optionally, choose your model, otherwise it will use the default Claude model via LLMGateway
4export ANTHROPIC_MODEL=gpt-5 # choose your model on llmgateway which supports tool calls
5
6# now run claude!
7claude

1export ANTHROPIC_BASE_URL=https://api.llmgateway.io
2export ANTHROPIC_AUTH_TOKEN=llmgtwy_.... # your llmgateway.io api key here
3# optionally, choose your model, otherwise it will use the default Claude model via LLMGateway
4export ANTHROPIC_MODEL=gpt-5 # choose your model on llmgateway which supports tool calls
5
6# now run claude!
7claude

Get Your API Key

1. Sign up free — no credit card required
2. Create a project and generate an API key
3. Start using it immediately (we provide model access, no provider keys needed)

Popular Models for Claude Code

Browse models with tool calling support. Here are popular choices:

Advanced Configuration

Model Switching

You can easily switch models by updating the environment variable:

1# Switch to GPT-4o Mini for cost savings
2export ANTHROPIC_MODEL=openai/gpt-4o-mini
3
4# Switch to Claude Sonnet for complex reasoning
5export ANTHROPIC_MODEL=anthropic/claude-3-5-sonnet-20241022

1# Switch to GPT-4o Mini for cost savings
2export ANTHROPIC_MODEL=openai/gpt-4o-mini
3
4# Switch to Claude Sonnet for complex reasoning
5export ANTHROPIC_MODEL=anthropic/claude-3-5-sonnet-20241022

To see the full list of models available to you, check out models with tool calls

Persistent Configuration

Add the environment variables to your shell profile (.bashrc, .zshrc, or .profile) for persistent configuration:

1echo 'export ANTHROPIC_BASE_URL=https://api.llmgateway.io' >> ~/.zshrc
2echo 'export ANTHROPIC_AUTH_TOKEN=llmgtwy_your_key_here' >> ~/.zshrc
3echo 'export ANTHROPIC_MODEL=openai/gpt-4o' >> ~/.zshrc
4source ~/.zshrc

1echo 'export ANTHROPIC_BASE_URL=https://api.llmgateway.io' >> ~/.zshrc
2echo 'export ANTHROPIC_AUTH_TOKEN=llmgtwy_your_key_here' >> ~/.zshrc
3echo 'export ANTHROPIC_MODEL=openai/gpt-4o' >> ~/.zshrc
4source ~/.zshrc

What You Get

Cost Visibility

Every Claude Code session is tracked in your dashboard. See exactly how many tokens you used, what it cost, and which model performed best. No more surprise bills.

Automatic Failover

If OpenAI goes down, LLM Gateway routes to a backup provider. Your coding session continues uninterrupted.

One Bill for Everything

Instead of managing credits across Anthropic, OpenAI, and Google, you get one account, one dashboard, one invoice.

Get Started

1. Sign up free — takes 30 seconds
2. Grab your API key from the dashboard
3. Set the three environment variables above
4. Run claude and start coding with any model

Questions? Check out our documentation or join our Discord.