List Models

from openai import OpenAI

client = OpenAI(
    base_url="https://api.parrotrouter.com/v1",
    api_key="YOUR_API_KEY"
)

# List all available models
models = client.models.list()

for model in models.data:
    print(f"Model: {model.id}")
    print(f"  Provider: {model.provider['name']}")
    print(f"  Context: {model.context_length} tokens")
    print(f"  Price: $" + str(model.pricing['prompt']) + "/1k prompt tokens")
    print(f"  Status: {model.status['message']}\n")

import requests

headers = {
    "Authorization": "Bearer YOUR_API_KEY"
}

# Get only chat models from OpenAI
response = requests.get(
    "https://api.parrotrouter.com/v1/models",
    headers=headers,
    params={
        "provider": "openai",
        "capability": "chat",
        "include_capabilities": True
    }
)

models = response.json()['data']

# Find models with function calling
function_calling_models = [
    model for model in models 
    if model['capabilities'].get('function_calling', False)
]

print(f"Found {len(function_calling_models)} models with function calling")

def select_best_model(models, requirements):
    """Select the best model based on requirements"""
    
    suitable_models = []
    
    for model in models['data']:
        # Check if model meets requirements
        if model['context_length'] < requirements['min_context']:
            continue
            
        if not all(model['capabilities'].get(cap, False) 
                  for cap in requirements['required_capabilities']):
            continue
            
        if not model['status']['active']:
            continue
            
        # Calculate score based on price and performance
        price_score = 1 / (model['pricing']['prompt'] + 
                          model['pricing']['completion'])
        perf_score = model['performance']['avg_tokens_per_second']
        
        suitable_models.append({
            'model': model,
            'score': price_score * perf_score
        })
    
    # Sort by score and return best
    suitable_models.sort(key=lambda x: x['score'], reverse=True)
    return suitable_models[0]['model'] if suitable_models else None

# Example usage
requirements = {
    'min_context': 8000,
    'required_capabilities': ['chat', 'function_calling'],
    'max_price_per_1k': 0.05
}

best_model = select_best_model(models, requirements)
print(f"Selected model: {best_model['id']}")