Skip to main content
The LLM Gateway API uses API keys to authenticate requests. You must include your API key in every request.

​
API Key Types

The gateway supports two authentication methods: Include your API key in the Authorization header: 
curl https://api.llmgateway.io/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

​
X-API-Key Header

Alternatively, use the x-api-key header: 
curl https://api.llmgateway.io/v1/chat/completions \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

​
Getting Your API Key

  1. Sign in to your LLM Gateway dashboard
  2. Navigate to the API Keys page
  3. Click Create API Key
  4. Copy and securely store your API key
API keys grant full access to your account. Never share them publicly or commit them to version control.

​
API Key Status

API keys can have the following statuses:
  • Active - Key is valid and can be used
  • Inactive - Key has been disabled
  • Deleted - Key has been permanently removed

​
Usage Limits

API keys can have optional usage limits: 
{
  "usageLimit": "1000.00",  // Maximum usage in USD
  "usage": "250.50"         // Current usage in USD
}
Once a key reaches its usage limit, requests will be rejected with a 401 Unauthorized error: 
{
  "error": true,
  "status": 401,
  "message": "Unauthorized: LLMGateway API key reached its usage limit."
}

​
Authentication Errors

​
Missing API Key

{
  "error": true,
  "status": 401,
  "message": "Unauthorized: No API key provided. Expected 'Authorization: Bearer your-api-token' header or 'x-api-key: your-api-token' header"
}

​
Invalid API Key

{
  "error": true,
  "status": 401,
  "message": "Unauthorized: Invalid LLMGateway API token. Please make sure the token is not deleted or disabled. Go to the LLMGateway 'API Keys' page to generate a new token."
}

​
Usage Limit Reached

{
  "error": true,
  "status": 401,
  "message": "Unauthorized: LLMGateway API key reached its usage limit."
}

​
Project Modes

API keys are associated with projects that operate in different modes:

​
API Keys Mode

Requires provider API keys to be configured in your project settings. The gateway uses your provider keys to make requests.

​
Credits Mode

Uses LLM Gateway’s shared pool of provider keys. Requests are billed against your account credits.

​
Hybrid Mode

Attempts to use your provider API keys first, then falls back to credits if no key is configured.

​
IAM Permissions

API keys can have IAM policies that restrict:
  • Which models can be accessed
  • Which providers can be used
  • Which operations are allowed
If an IAM policy denies access, you’ll receive a 403 Forbidden error: 
{
  "error": true,
  "status": 403,
  "message": "Access denied: No providers are allowed for model gpt-4 after applying IAM rules."
}

​
Best Practices

Create new API keys periodically and delete old ones to minimize security risks.
Store API keys in environment variables instead of hardcoding them:
export LLMGATEWAY_API_KEY="your-api-key"
Configure usage limits on API keys to prevent unexpected charges from runaway processes.
Create different API keys for development, staging, and production environments.

​
Testing Authentication

Test your API key with the models endpoint: 
curl https://api.llmgateway.io/v1/models \
  -H "Authorization: Bearer YOUR_API_KEY"
If authentication succeeds, you’ll receive a list of available models.

Build docs developers (and LLMs) love

Get started for freeTalk to us