Skip to content

response.md

Latest commit

 

History

History
155 lines (117 loc) · 5.69 KB

response.md

File metadata and controls

155 lines (117 loc) · 5.69 KB
Raw
page_title openai_response Resource - terraform-provider-openai
subcategory
description Generates a response using the OpenAI Responses API.

openai_response (Resource)

Generates a response using the OpenAI Responses API.

Example Usage

resource "openai_response" "full_example" {
  model = "gpt-5.2"
  input = "What is the weather in San Francisco?"

  # Context
  instructions = "You are a helpful assistant."
  # conversation_id = "conv_12345" # Optional: To continue a conversation

  # Tuning
  top_p             = 1.0
  max_output_tokens = 100
  truncation        = "auto"
  reasoning_effort  = "medium"

  # Metadata
  metadata = {
    "environment" = "test"
    "user_id"     = "user_123"
  }

  # Tools
  tools = [
    {
      type = "function"
      function = {
        name        = "get_weather"
        description = "Get the current weather in a given location"
        parameters = jsonencode({
          type = "object"
          properties = {
            location = {
              type        = "string"
              description = "The city and state, e.g. San Francisco, CA"
            }
            unit = {
              type = "string"
              enum = ["celsius", "fahrenheit"]
            }
          }
          required = ["location"]
        })
      }
    },
    {
      type = "web_search"
    }
  ]
  tool_choice         = "auto"
  parallel_tool_calls = true

  # Output Format
  response_format = "text" # or json_object, json_schema
  include         = ["web_search_call.action.sources"]
}

output "full_example_output" {
  value = openai_response.full_example.content
}

Schema

Required

  • input (String) The input text for the response.
  • model (String) The model to use for the response.

Optional

  • conversation_id (String) The unique ID of the conversation to initiate or continue.
  • include (List of String) Specify additional output data to include in the model response. Currently supported values include web_search_call.action.sources, code_interpreter_call.outputs, etc.
  • instructions (String) A system (or developer) message inserted into the model's context.
  • max_output_tokens (Number) The maximum number of tokens to generate in the response.
  • max_tool_calls (Number) The maximum number of tool calls to make in the response.
  • metadata (Map of String) Set of key-value pairs that can be attached to an object. This can be useful for storing additional information about the object in a structured format.
  • parallel_tool_calls (Boolean) Whether to allow parallel tool calls. Defaults to true.
  • previous_response_id (String) The unique ID of the previous response to the model. Use this to create multi-turn conversations.
  • prompt (Attributes) Reference to a prompt template and its variables. (see below for nested schema)
  • reasoning_effort (String) Constrains effort on reasoning for reasoning models. Valid values are low, medium, high.
  • response_format (String) Specifies the format that the model must output. Compatible with json_object, json_schema, text.
  • temperature (Number) What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic.
  • tool_choice (String) Controls which (if any) tool is called by the model. Can be none, auto, required, or a specific function name.
  • tools (Attributes List) A list of tools the model may call. Currently, only functions are supported as a tool. (see below for nested schema)
  • top_logprobs (Number) An integer between 0 and 20 specifying the number of most likely tokens to return at each token position.
  • top_p (Number) An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass.
  • truncation (String) Controls how the model truncates the context if it exceeds the maximum token limit. Valid values: auto, disabled.

Read-Only

  • content (String) The concatenated text content of the response. This is a convenience attribute for easy access to the generated text.
  • created_at (Number) The Unix timestamp (in seconds) of when the response was created.
  • id (String) The ID of the generated response.
  • output (Attributes List) The generated output items. (see below for nested schema)

Nested Schema for prompt

Required:

  • id (String) The unique ID of the prompt template to use.

Optional:

  • variables (String) JSON string of variables to pass to the prompt template.
  • version (String) Optional version of the prompt template.

Nested Schema for tools

Required:

  • type (String) The type of the tool. Common values: function, web_search, file_search, computer_use, code_interpreter.

Optional:

Nested Schema for tools.function

Required:

  • name (String) The name of the function to be called.
  • parameters (String) The parameters the functions accepts, described as a JSON Schema object.

Optional:

  • description (String) A description of what the function does, used by the model to choose when and how to call the function.

Nested Schema for output

Read-Only:

  • content (String) The content of the output item. Currently only text content is extracted.
  • type (String)