OpenRouter Provider

The OpenRouter provider enables access to multiple LLM models through the OpenRouter API, providing a unified interface to various AI models from different providers.

Key Features

• Multi-model access: Single API endpoint for multiple LLM providers
• Cost transparency: Clear pricing information for all supported models
• Standardized API: Consistent interface across different model providers
• Usage tracking: Built-in token counting and cost calculation
• Model switching: Easily switch between different models without changing providers

Configuration

from aicore.llm.config import LlmConfig

config = LlmConfig(
    provider="openrouter",
    api_key="your_openrouter_api_key",  # Get from https://openrouter.ai/keys
    model="openai/gpt-4o",  # Format: provider/model-name
    temperature=0.7,
    max_tokens=1000
)

Supported Models

OpenRouter supports models from multiple providers, for a complete list of available models, see the OpenRouter Models page.

Usage Examples

Basic Completion

from aicore.llm import Llm

llm = Llm(config=config)
response = llm.complete("Explain quantum computing in simple terms")
print(response.content)

With Message History

messages = [
    {"role": "system", "content": "You are a helpful assistant"},
    {"role": "user", "content": "What's the weather today?"}
]

response = llm.chat(messages)

Observability Integration

All OpenRouter operations are automatically tracked by the observability system:

from aicore.observability.collector import LlmOperationCollector

# View OpenRouter operations
operations = LlmOperationCollector.polars_from_db(provider="openrouter")
print(operations.select(["model", "latency_ms", "total_tokens"]))

Troubleshooting

• Authentication Errors: Verify your API key at OpenRouter Keys
• Model Not Found: Check the exact model name format (provider/model-name)
• Rate Limits: OpenRouter has default rate limits - consider upgrading if needed