Parameter Validation¶

This guide explains how to validate parameters against model constraints using the OpenAI Model Registry.

Why Validate Parameters?¶

Parameter validation ensures that your application uses valid values for model parameters, preventing runtime errors when calling the OpenAI API. Different models may have different constraints for the same parameter.

Basic Parameter Validation¶

You can validate parameters through the model capabilities object:

from openai_model_registry import ModelRegistry

registry = ModelRegistry.get_default()
capabilities = registry.get_capabilities("gpt-4o")

# Validate a temperature value
try:
    capabilities.validate_parameter("temperature", 0.7)
    print("✅ Valid temperature")
except ValueError as e:
    print(f"❌ Invalid temperature: {e}")
# Expected output: ✅ Valid temperature

Handling Validation Errors¶

Validation errors provide detailed information about why a parameter is invalid:

from openai_model_registry import ModelRegistry

registry = ModelRegistry.get_default()
capabilities = registry.get_capabilities("gpt-4o")

try:
    # Invalid temperature (outside of allowed range)
    capabilities.validate_parameter("temperature", 3.0)
except Exception as e:
    print(f"Validation error: {e}")
    # Expected output: Validation error: Parameter 'temperature' must be between 0 and 2.
    # Description: Controls randomness: Lowering results in less random completions.
    # Current value: 3.0

Common Parameter Types¶

Numeric Parameters¶

Numeric parameters typically have constraints for:

Minimum value
Maximum value
Whether the parameter allows floats, integers, or both

Examples of numeric parameters:

temperature: Controls randomness (typically 0-2)
top_p: Controls diversity via nucleus sampling (typically 0-1)
max_tokens: Controls maximum completion length (typically 1-model_max)

Enum Parameters¶

Enum parameters accept only specific string values from a predefined list.

Examples of enum parameters:

response_format: Format of the model's output (e.g., "text", "json_schema")
reasoning_effort (O1 model): Level of reasoning effort (e.g., "low", "medium", "high")

Model-Specific Parameters¶

Different models may support different parameters. For example, the O1 model has parameters not available in other models:

from openai_model_registry import ModelRegistry

# Initialize registry
registry = ModelRegistry.get_default()

# Get O1 capabilities
o1_capabilities = registry.get_capabilities("o1")

# Validate O1-specific parameter
try:
    o1_capabilities.validate_parameter("reasoning_effort", "medium")
    print("reasoning_effort 'medium' is valid for O1")
except Exception as e:
    print(f"Invalid parameter: {e}")
# Expected output: reasoning_effort 'medium' is valid for O1

Getting Supported Parameters¶

You can retrieve the list of parameters supported by a specific model:

from openai_model_registry import ModelRegistry

registry = ModelRegistry.get_default()
capabilities = registry.get_capabilities("gpt-4o")

# Get supported parameters
supported_params = [ref.ref.split(".")[1] for ref in capabilities.supported_parameters]
print(f"Supported parameters: {', '.join(sorted(supported_params))}")
# Expected output: Supported parameters: max_tokens, temperature, top_p, ...

Next Steps¶

Now that you understand parameter validation, learn about Advanced Usage for more complex scenarios and registry management.