AeonSage supports 20+ AI providers, giving you the flexibility to use the best model for your needs. Switch between providers seamlessly without changing your agent code.
Supported Providers
Cloud Providers
Local Providers
Aggregators
Provider Architecture
┌─────────────────────────────────────────────────────────────────┐
│ Provider Selection │
│ │
│ ┌──────────────────────────────────────────────────────────┐ │
│ │ Gateway Router │ │
│ │ │ │ │
│ │ ┌───────────────┼───────────────┐ │ │
│ │ ▼ ▼ ▼ │ │
│ │ ┌──────────┐ ┌──────────┐ ┌──────────┐ │ │
│ │ │ Primary │ │ Fallback │ │ Local │ │ │
│ │ │ Provider │ │ Provider │ │ Provider │ │ │
│ │ │ (OpenAI) │ │(Anthropic)│ │ (Ollama) │ │ │
│ │ └──────────┘ └──────────┘ └──────────┘ │ │
│ │ │ │ │ │ │
│ │ └───────────────┼───────────────┘ │ │
│ │ ▼ │ │
│ │ ┌──────────────────┐ │ │
│ │ │ Unified Response │ │ │
│ │ └──────────────────┘ │ │
│ └──────────────────────────────────────────────────────────┘ │
│ │
└─────────────────────────────────────────────────────────────────┘
Configuration
Environment Variables
Configuration File
{
"providers": {
"default": "openai",
"openai": {
"apiKey": "${OPENAI_API_KEY}",
"model": "gpt-4o",
"temperature": 0.7
},
"anthropic": {
"apiKey": "${ANTHROPIC_API_KEY}",
"model": "claude-3-5-sonnet-latest"
},
"ollama": {
"host": "http://localhost:11434",
"model": "llama3.2"
}
}
}
Provider Selection
Default Provider
Set the default provider for all requests:
aeonsage config set providers.default openai
Per-Channel Provider
Different channels can use different providers:
{
"channels": {
"telegram": {
"provider": "openai",
"model": "gpt-4o"
},
"discord": {
"provider": "anthropic",
"model": "claude-3-5-sonnet-latest"
}
}
}
Fallback Chain
Configure automatic fallback when primary fails:
{
"providers": {
"fallback": ["openai", "anthropic", "ollama"],
"fallbackDelay": 5000
}
}
Model Parameters
Common Parameters
Configuration Example
{
"providers": {
"openai": {
"model": "gpt-4o",
"parameters": {
"temperature": 0.7,
"maxTokens": 4096,
"topP": 0.9,
"frequencyPenalty": 0.3
}
}
}
}
Testing Providers
Connection Test
# Test all providers
aeonsage providers test
# Test specific provider
aeonsage providers test openai
# Test with custom message
aeonsage providers test openai --message "Hello, world!"
# Run benchmark
aeonsage providers benchmark
# Compare providers
aeonsage providers benchmark openai anthropic
Cost Optimization
Cost Tracking
# View usage statistics
aeonsage providers stats
# View cost breakdown
aeonsage providers costs --period month
Optimization Strategies
Troubleshooting
Common Issues
Next Steps
