Supported Providers

Overview

AI Gateway supports multiple LLM providers through a unified OpenAI-compatible API. Add your provider API keys in the dashboard to enable access to their models.

Provider Matrix

Adding Provider Keys

1. Navigate to AI Gateway Settings
2. Under "Provider Keys", click Add Key
3. Select your provider from the dropdown
4. Paste your API key
5. Click Save

Key Security

• Provider keys are encrypted at rest
• Keys are never exposed in logs or responses
• You can rotate keys at any time
• Deleting a key immediately revokes access

Model Naming

Use the provider's model names directly:

// OpenAI
model: 'gpt-4o'
model: 'gpt-4-turbo'
model: 'gpt-3.5-turbo'
model: 'o1'
model: 'o1-mini'
 
// Anthropic
model: 'claude-3-5-sonnet'
model: 'claude-3-opus'
model: 'claude-3-haiku'
model: 'claude-sonnet-4'
model: 'claude-opus-4'

Provider Priority

When you have multiple providers configured, you can set priority for fallback:

1. Go to Settings > Provider Keys
2. Drag providers to reorder priority
3. The first provider is used by default
4. If it fails, the next provider is tried

Provider-Specific Limits

These limits are from the provider. AI Gateway respects and passes through provider rate limits.

Using Multiple Providers

Fallback Configuration

Set up automatic failover between providers:

// Primary request goes to OpenAI
// If it fails, AI Gateway tries Anthropic
const response = await openai.chat.completions.create({
  model: 'gpt-4o', // Primary
  messages: [...],
});

Configure fallback mapping in Settings:

Load Balancing

Coming soon: Distribute requests across providers based on cost, latency, or availability.

Checking Provider Status

View provider health in the dashboard:

1. Go to AI Gateway > Analytics
2. See real-time status for each provider
3. View error rates and latency by provider

Next Steps

• OpenAI Integration - Full OpenAI setup guide
• Anthropic Integration - Full Anthropic setup guide
• Fallback Configuration - Set up automatic failover