Create response

POSThttps://api.poe.com/v1/responses

Overview

Creates a model response using the OpenAI Responses API format. This is the most feature-rich endpoint, supporting advanced capabilities beyond basic chat completions.

Key features:

Reasoning and extended thinking support
Built-in tool calling with web search
Structured outputs (JSON schema)
Streaming support via SSE
Multi-turn conversations via previous_response_id
Multi-modal inputs (text, images)

OpenAI Responses API compatible: This endpoint follows the OpenAI Responses API format. You can use the OpenAI SDK with Poe's base URL as a drop-in replacement.

Note: This endpoint supports all models available on Poe, not just OpenAI models.

Authentication

Send your Poe API key in the Authorization header:

Authorization: Bearer sk_test_51SAMPLEKEY

All requests must be made over HTTPS.

Parameters

This endpoint does not accept query or path parameters.

Request body

Field	Type	Required	Description
`model`	string	Required	ID of the model to use. Use Poe bot names. Note: Poe UI-specific system prompts are skipped
`input`	string \| object[]	Required	The input to the model. Can be a simple string or an array of input items including text, images, and previous assistant messages.
`instructions`	string \| null	Optional	A system (or developer) message inserted at the beginning of the model's context
`stream`	boolean	Optional	Whether to stream back partial progress as server-sent events (SSE) Default: `false`
`max_output_tokens`	integer \| null	Optional	Maximum number of output tokens to generate. Uses the model's default if not specified
`temperature`	number \| null	Optional	Sampling temperature between 0 and 2 Min: 0 · Max: 2
`top_p`	number \| null	Optional	Nucleus sampling parameter Min: 0 · Max: 1
`reasoning`	object \| null	Optional	Configuration for reasoning/thinking models. Not all models support reasoning.
`text`	object \| null	Optional	Configuration for structured text output
`tools`	array \| null	Optional	List of tools the model may call. Supports function tools and built-in tools like `web_search_preview`.
`tool_choice`	string \| object	Optional	How the model should select which tool to use. Can be `"auto"`, `"required"`, `"none"`, or a specific tool object.
`parallel_tool_calls`	boolean \| null	Optional	Whether to allow the model to run tool calls in parallel
`truncation`	string \| null	Optional	The truncation strategy to use. - `"auto"`: Truncate input to fit the model's context window - `"disabled"`: Fail if the input exceeds the context window Allowed values: `auto`, `disabled`
`previous_response_id`	string \| null	Optional	The ID of a previous response for multi-turn conversations. The model will use the previous response's context
`include`	array \| null	Optional	Additional output data to include in the response. Supported values: - `"web_search_call.action.sources"` - Include web search sources - `"message.output_text.logprobs"` - Include log probabilities - `"reasoning.encrypted_content"` - Include encrypted reasoning content
`metadata`	object \| null	Optional	Set of key-value pairs for storing additional information (keys max 64 chars, values max 512 chars)
`service_tier`	string \| null	Optional	Specifies the processing tier for the request Allowed values: `auto`, `default`, `flex`, `priority`
`store`	boolean \| null	Optional	Whether to store the generated response for later retrieval

Responses

Field	Type	Required	Description
`id`	string	Optional	Unique identifier for the response
`object`	"response"	Optional	Object type, always "response" Allowed values: `response`
`created_at`	integer	Optional	Unix timestamp of when the response was created
`model`	string	Optional	The model used to generate the response
`output`	object[]	Optional	Array of output items generated by the model
`output[].type`	string	Optional	Type of output item (e.g., "message")
`output[].role`	string	Optional	Role of the output (e.g., "assistant")
`output[].content`	object[]	Optional	Content blocks in the output
`output[].content[].type`	string	Optional	Content block type (e.g., "output_text")
`output[].content[].text`	string	Optional	Text content
`status`	"completed" \| "failed" \| "in_progress" \| "incomplete"	Optional	The status of the response Allowed values: `completed`, `failed`, `in_progress`, `incomplete`
`usage`	object	Optional	Token usage information
`usage.input_tokens`	integer	Optional	Number of input tokens
`usage.output_tokens`	integer	Optional	Number of output tokens
`usage.total_tokens`	integer	Optional	Total number of tokens used

❌ Error codes

Http	Type	Description
400	`invalid_request_error`	Bad request Malformed JSON or missing required fields
401	`authentication_error`	Authentication failed Invalid API key
402	`insufficient_credits`	Insufficient credits Point balance is zero or negative
429	`rate_limit_error`	Rate limit exceeded Rate limit exceeded (500 requests per minute)

🔁 Callbacks & webhooks

No callbacks or webhooks are associated with this endpoint.

Code example

Create a model response with advanced features like reasoning, tools, and web search.

curl https://api.poe.com/v1/responses \
-X POST \
-H "Authorization: Bearer sk_test_51SAMPLEKEY" \
-H "Accept: application/json" \
-H "Content-Type: application/json" \
-d '{
  "model": "Claude-Sonnet-4.5",
  "input": "What are the top 3 things to do in New York?"
}'

Normal response

{
"id": "resp_abc123",
"object": "response",
"created_at": 1704825600,
"model": "Claude-Sonnet-4.5",
"output": [
  {
    "type": "message",
    "role": "assistant",
    "content": [
      {
        "type": "output_text",
        "text": "Here are the top 3 things to do in New York:\n\n1. Visit Central Park\n2. See the Statue of Liberty\n3. Experience Times Square"
      }
    ]
  }
],
"status": "completed",
"usage": {
  "input_tokens": 15,
  "output_tokens": 45,
  "total_tokens": 60
}
}

Streaming response

data: {"type":"response.output_text.delta","delta":"Here"}

data: {"type":"response.output_text.delta","delta":" are"}

data: {"type":"response.output_text.delta","delta":" the top"}

data: {"type":"response.completed","response":{"id":"resp_abc123","object":"response","status":"completed","usage":{"input_tokens":15,"output_tokens":45,"total_tokens":60}}}

data: [DONE]