Skip to main content

​
Parameters

ParameterTypeDefaultDescription
supports_toolsboolTrueWhether the model supports tools
supports_json_schema_outputboolTrueWhether the model supports JSON schema output
supports_json_object_outputboolTrueWhether the model supports JSON object output
default_structured_output_modeStructuredOutputMode'tool'The default structured output mode to use for the model
prompted_output_templatestr"Always respond with a JSON object that's compatible with this schema:\n\n{schema}\n\nDon't include any text or Markdown fencing before or after.\n"The instructions template to use for prompted structured output. The '' placeholder will be replaced with the JSON schema for the output.
json_schema_transformertype[JsonSchemaTransformer] | NoneOpenAIJsonSchemaTransformerThe transformer to use to make JSON schemas for tools and structured output compatible with the model
thinking_tagstuple[str, str]('<think>', '</think>')The tags used to indicate thinking parts in the model’s output. Defaults to ('', '').
ignore_streamed_leading_whitespaceboolFalseWhether to ignore leading whitespace when streaming a response
openai_supports_strict_tool_definitionboolTrueThis can be set by a provider or user if the OpenAI-”compatible” API doesn’t support strict tool definitions
openai_supports_sampling_settingsboolTrueTurn off to don’t send sampling settings like temperature and top_p to models that don’t support them, like OpenAI’s o-series reasoning models
openai_unsupported_model_settingsSequence[str]()A list of model settings that are not supported by the model
openai_supports_tool_choice_requiredboolTrueWhether the provider accepts the value tool_choice='required' in the request payload
openai_system_prompt_roleOpenAISystemPromptRole | NoneNoneThe role to use for the system prompt message. If not provided, defaults to 'system'
openai_chat_supports_web_searchboolFalseWhether the model supports web search in Chat Completions API
openai_supports_encrypted_reasoning_contentboolFalseWhether the model supports including encrypted reasoning content in the response

​
Functions

​
openai_model_profile

Get the model profile for an OpenAI model. Parameters:
  • model_name (str): The name of the OpenAI model
Returns:
  • ModelProfile: The model profile for the OpenAI model
Description: This function returns the model profile for an OpenAI model. It creates an OpenAIModelProfile instance with specific configurations based on the model name. For reasoning models (starting with ‘o’ or ‘gpt-5’), it disables certain sampling settings. For search-preview models, it enables web search support. For o1-mini models, it sets the system prompt role to ‘user’.

​
OpenAIJsonSchemaTransformer

Recursively handle the schema to make it compatible with OpenAI strict mode. Description: This transformer ensures compatibility with OpenAI’s strict mode requirements:
  • additionalProperties must be set to false for each object in the parameters
  • All fields in properties must be marked as required
  • Handles various schema transformations and optimizations
Parameters:
  • schema (JsonSchema): The JSON schema to transform
  • strict (bool | None): Whether to use strict mode
Returns:
  • JsonSchema: The transformed JSON schema compatible with OpenAI strict mode
⌘I