🎯 Smart Model Selection

Method: smart_select()

OmniRouter provides a unique smart model selection feature that automatically chooses the best model for your specific needs by balancing factors like accuracy, cost, and latency.

📝 Example Usage

from omnilabs import OmniClient, ChatMessage

client = OmniClient()
response = client.smart_select(
    messages=[ChatMessage(role="user", content="Solve this calculus problem: ∫x²dx")],
    rel_accuracy=0.8,  # Prioritize accuracy
    rel_cost=0.2,
    verbose=True
)
print(f"Selected model: {response['model']}")
print(f"Response: {response['content']}")

🔹Parameters

Parameter	Type	Description
messages	List[Union[Dict[str, str], ChatMessage]]	The chat messages to analyze for model selection
k	int	Number of top models to consider (default: 5)
model_names	Optional[List[str]]	Specific models to select from (default: None)
rel_cost	float	Relative importance of cost (0-1) (default: 0.5)
rel_latency	float	Relative importance of latency (0-1) (default: 0.0)
rel_accuracy	float	Relative importance of accuracy (0-1) (default: 0.5)
verbose	bool	Whether to return detailed explanation (default: False)

🔹Response Format

{
    "model": "gpt-4",
    "content": "Generated response",
    "explanation": "Selected GPT-4 due to high accuracy requirements...",
    "provider": "openai",
    "usage": {
        "all_tokens": 45,
        "total_cost": 0.00002
    }
}

The smart selection algorithm considers:

• Accuracy: How well the model performs on complex tasks
• Cost: Token pricing and efficiency
• Latency: Response time for generation

You can adjust these priorities using the relative importance parameters (rel_cost, rel_latency, rel_accuracy).

---