Responses endpoint - return JSON does not match spec

[RichNasz]

System Info

• Development on macOS using Xcode and using the the SwiftOpenAI OSS project for the client SDK that follows the OpenAI spec closely.
• Latest Llama Stack release

Information

• The official example scripts
• My own modified scripts

🐛 Describe the bug

Right now I'm getting an error processing data coming back from the endpoint as it appears LLS isn't returning a complete Response object per the OpenAI platform documentation which is causing a JSON decoding error. The missing "tool_choice" return value is breaking my code now, but there are other fields not being returned as well.

Here is the curl for the request I made:
curl http://localhost:8321/v1/openai/v1/responses -X POST
-H "Content-Type: application/json"
-H "Authorization: Bea................er "
-d '{
"model" : "ollama/llama3.2:3b",
"stream" : false,
"input" : "Say hello!"
}'

Here is the response:
{"created_at":1754053317,"error":null,"id":"resp-c488d856-c083-47c1-9a67-e852ccc61e15","model":"ollama/llama3.2:3b","object":"response","output":[{"content":[{"text":"Hello! It's nice to meet you. Is there something I can help you with, or would you like to chat for a bit?","type":"output_text","annotations":[]}],"role":"assistant","type":"message","id":"msg_10a12839-ea2e-4739-bf79-90ec8dce0c5f","status":"completed"}],"parallel_tool_calls":false,"previous_response_id":null,"status":"completed","temperature":null,"text":{"format":{"type":"text"}},"top_p":null,"truncation":null,"user":null}

Error logs

Error is a result of JSON decoder missing required fields and associated values.

Expected behavior

That an LLS response to the /v1/openai/v1/responses endpoint will return a data compliant with the OpenAI spec.

[nathan-weinberg]

Could you share the JSON of the response you expect? As well as a link to the spec?

[RichNasz]

The link to the spec if at this URL: https://platform.openai.com/docs/api-reference/responses/create

When you open that web page, on the right you will see 5 different Responses requests along with the associated response JSON. The detailed documentation for the response JSON is included on the web page as well.

For example, here is a sample of the JSON structure the spec indicates should have been provided when I made the simple text request provided in the description of this issue:

{
"id": "resp_67ccf4c55fc48190b71bd0463ad3306d09504fb6872380d7",
"object": "response",
"created_at": 1741485253,
"status": "completed",
"error": null,
"incomplete_details": null,
"instructions": null,
"max_output_tokens": null,
"model": "gpt-4.1-2025-04-14",
"output": [
{
"type": "file_search_call",
"id": "fs_67ccf4c63cd08190887ef6464ba5681609504fb6872380d7",
"status": "completed",
"queries": [
"attributes of an ancient brown dragon"
],
"results": null
},
{
"type": "message",
"id": "msg_67ccf4c93e5c81909d595b369351a9d309504fb6872380d7",
"status": "completed",
"role": "assistant",
"content": [
{
"type": "output_text",
"text": "The attributes of an ancient brown dragon include...",
"annotations": [
{
"type": "file_citation",
"index": 320,
"file_id": "file-4wDz5b167pAf72nx1h9eiN",
"filename": "dragons.pdf"
},
{
"type": "file_citation",
"index": 576,
"file_id": "file-4wDz5b167pAf72nx1h9eiN",
"filename": "dragons.pdf"
},
{
"type": "file_citation",
"index": 815,
"file_id": "file-4wDz5b167pAf72nx1h9eiN",
"filename": "dragons.pdf"
},
{
"type": "file_citation",
"index": 815,
"file_id": "file-4wDz5b167pAf72nx1h9eiN",
"filename": "dragons.pdf"
},
{
"type": "file_citation",
"index": 1030,
"file_id": "file-4wDz5b167pAf72nx1h9eiN",
"filename": "dragons.pdf"
},
{
"type": "file_citation",
"index": 1030,
"file_id": "file-4wDz5b167pAf72nx1h9eiN",
"filename": "dragons.pdf"
},
{
"type": "file_citation",
"index": 1156,
"file_id": "file-4wDz5b167pAf72nx1h9eiN",
"filename": "dragons.pdf"
},
{
"type": "file_citation",
"index": 1225,
"file_id": "file-4wDz5b167pAf72nx1h9eiN",
"filename": "dragons.pdf"
}
]
}
]
}
],
"parallel_tool_calls": true,
"previous_response_id": null,
"reasoning": {
"effort": null,
"summary": null
},
"store": true,
"temperature": 1.0,
"text": {
"format": {
"type": "text"
}
},
"tool_choice": "auto",
"tools": [
{
"type": "file_search",
"filters": null,
"max_num_results": 20,
"ranking_options": {
"ranker": "auto",
"score_threshold": 0.0
},
"vector_store_ids": [
"vs_1234567890"
]
}
],
"top_p": 1.0,
"truncation": "disabled",
"usage": {
"input_tokens": 18307,
"input_tokens_details": {
"cached_tokens": 0
},
"output_tokens": 348,
"output_tokens_details": {
"reasoning_tokens": 0
},
"total_tokens": 18655
},
"user": null,
"metadata": {}
}

[mattf]

	openai.types.responses.Response	llama_stack.apis.agents.OpenAIResponseObject
id	str	str
created_at	float	int
error	Optional[ResponseError]	Optional[OpenAIResponseError]
incomplete_details	Optional[IncompleteDetails]	—
instructions	Union[str, List[ResponseInputItem], None]	—
metadata	Optional[Metadata]	—
model	ResponsesModel	str
object	Literal["response"]	Literal["response"]
output	List[ResponseOutputItem]	list[OpenAIResponseOutput]
parallel_tool_calls	bool	bool
temperature	Optional[float]	Optional[float]
tool_choice	ToolChoice	—
tools	List[Tool]	—
top_p	Optional[float]	Optional[float]
background	Optional[bool]	—
max_output_tokens	Optional[int]	—
max_tool_calls	Optional[int]	—
previous_response_id	Optional[str]	Optional[str]
prompt	Optional[ResponsePrompt]	—
prompt_cache_key	Optional[str]	—
reasoning	Optional[Reasoning]	—
safety_identifier	Optional[str]	—
service_tier	Optional[Literal["auto", "default", "flex", "scale", "priority"]]	—
status	Optional[ResponseStatus]	str
text	Optional[ResponseTextConfig]	OpenAIResponseText
top_logprobs	Optional[int]	—
truncation	Optional[Literal["auto", "disabled"]]	Optional[str]
usage	Optional[ResponseUsage]	—
user	Optional[str]	Optional[str]

[iamemilio]

PR in flight to patch this 👍 . As a first pass, I will implement these fields by hand and link that PR here. I want to make a follow up to this to see if the openAPI fields will be preserved if we replace the OpenAIResponseObject with the OpenAI python object as well.

[github-actions]

This issue has been automatically marked as stale because it has not had activity within 60 days. It will be automatically closed if no further activity occurs within 30 days.