curl --request POST \
--url https://api.pre.dev/v1/responses \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"model": "deepseek/deepseek-v4.1-flash",
"input": "Say hello in five words."
}
'{
"id": "pdg-043293a6-d1a8-4695-81fe-603ad4661ebb",
"object": "response",
"created_at": 1790500000,
"model": "deepseek/deepseek-v4.1-flash",
"status": "completed",
"output": [
{
"type": "message",
"role": "assistant",
"content": [
{
"type": "output_text",
"text": "Hello there, nice to meet you."
}
]
}
],
"usage": {
"input_tokens": 12,
"output_tokens": 8,
"total_tokens": 20
}
}Responses
Requests in the OpenAI Responses format. Each request stands alone, so send the whole conversation.
curl --request POST \
--url https://api.pre.dev/v1/responses \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"model": "deepseek/deepseek-v4.1-flash",
"input": "Say hello in five words."
}
'{
"id": "pdg-043293a6-d1a8-4695-81fe-603ad4661ebb",
"object": "response",
"created_at": 1790500000,
"model": "deepseek/deepseek-v4.1-flash",
"status": "completed",
"output": [
{
"type": "message",
"role": "assistant",
"content": [
{
"type": "output_text",
"text": "Hello there, nice to meet you."
}
]
}
],
"usage": {
"input_tokens": 12,
"output_tokens": 8,
"total_tokens": 20
}
}Authorizations
The workspace API key (pdk_…) from Integrations → Built-in in the pre.dev dashboard.
Headers
Attribute this call to one of the workspace's projects. An id the workspace does not own is ignored.
Body
"deepseek/deepseek-v4.1-flash"
A string, or an array of input items (messages, function call outputs, and so on). Send the whole conversation: responses are not stored, and previous_response_id is not supported.
x >= 1Output format, for example { "format": { "type": "json_schema", … } }.
Reasoning controls for models that support them. Reasoning tokens count as output tokens.
Show child attributes
Show child attributes
Response
The response object, or Responses API events when stream is true.
Generation id (pdg-…).
"response"completed, incomplete, in_progress, failed, cancelled, queued Output items: messages with output_text parts, reasoning, and function calls.
The concatenated text output, when present.
Show child attributes
Show child attributes
Was this page helpful?

