/Docs
Back to dashboard

Claude Code with Free $10 API Usage: LLMsRelay Setup

Create an LLMsRelay key, claim $10 in promotional API usage, and configure a Claude-compatible client with a custom base URL.

LLMsRelay is an independently operated API gateway. It is not affiliated with or endorsed by Anthropic, PBC. Product and model names are used only to describe compatibility.

Start with $10 in Promotional API Usage

LLMsRelay provides its own API keys for an independently operated gateway. You can claim $10 in promotional API usage through the dashboard reward flow and use a compatible custom base URL in your local development tools. Check GET /v1/models before choosing a model ID.

This is not a claude.ai account or subscription. Use of Claude Code and other third-party tools remains subject to their respective terms and version-specific compatibility.
Get $10 free usage

How to Configure a Compatible Client

Use this flow to test your LLMsRelay key with a local client that supports a custom base URL:

  1. Create an LLMsRelay key in the dashboard and claim the reward if eligible.
  2. Confirm model IDs with GET /v1/models.
  3. Install and configure your local client according to its documentation.
  4. Set the gateway endpoint and your LLMsRelay key in a test environment.
  5. Run a small request first before using the setup in a larger project.
Example environment configurationbash
# Install Claude Code using its official installation instructions
npm install -g @anthropic-ai/claude-code

# Set your LLMsRelay API key
export ANTHROPIC_BASE_URL="https://api.llmsrelay.com"
export ANTHROPIC_API_KEY="YOUR_SK_CS4_KEY"
export ANTHROPIC_MODEL="MODEL_ID_FROM_V1_MODELS"
export ANTHROPIC_SMALL_FAST_MODEL="FAST_MODEL_ID_FROM_V1_MODELS"

# Launch the local client
claude

What to Verify Before Use

CheckWhy it matters
Custom base URL supportThe local client must allow the configured endpoint
Model IDUse GET /v1/models instead of assuming a static model list
Small test requestConfirms your local configuration and network path
Usage limitPrevents a test session from consuming more than intended
Client updatesA future client release can change custom gateway behavior

Choosing a Model for Coding Tasks

  • Smaller or faster routes for short edits, test generation, and simple automation.
  • Balanced routes for daily coding tasks and analysis.
  • Larger routes for complex refactoring, broad code changes, and architecture work.
Select a returned model IDbash
# Use a model ID returned by GET /v1/models
claude --model MODEL_ID_FROM_V1_MODELS

# Test a second available model when your workflow needs it
claude --model ANOTHER_MODEL_ID_FROM_V1_MODELS

Frequently Asked Questions

Can I start Claude Code with promotional API usage?

You can create an LLMsRelay key, claim $10 in promotional API usage through the dashboard reward flow, and configure a compatible base URL where your local client supports it.

Is this an Anthropic subscription or account?

No. LLMsRelay is independently operated and provides its own API keys. It does not provide claude.ai accounts or subscriptions.

How do I confirm the available models?

Call GET /v1/models with your LLMsRelay key before configuration. It returns the model IDs currently available to that key.

Will every Claude Code feature work through a custom gateway?

Compatibility depends on the client version, the configured request format, and the selected model route. Test your intended workflow with a non-production key before rollout.

Ready to start?

Create a key and configure a compatible API route in under 2 minutes.

View Plans
PreviousSupport & RefundsNext Free Claude API Key