phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-07-13 16:46:09 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

feat: Structured output for Responses API (#2324)

Browse source 
# What does this PR do?

This adds the missing `text` parameter to the Responses API that is how
users control structured outputs. All we do with that parameter is map
it to the corresponding chat completion response_format.

## Test Plan

The new unit tests exercise the various permutations allowed for this
property, while a couple of new verification tests actually use it for
real to verify the model outputs are following the format as expected.

Unit tests:

`python -m pytest -s -v
tests/unit/providers/agents/meta_reference/test_openai_responses.py`

Verification tests:

```
llama stack run llama_stack/templates/together/run.yaml
pytest -s -vv 'tests/verifications/openai_api/test_responses.py' \
  --base-url=http://localhost:8321/v1/openai/v1 \
  --model meta-llama/Llama-4-Scout-17B-16E-Instruct
```

Note that the verification tests can only be run with a real Llama Stack
server (as opposed to using the library client via
`--provider=stack:together`) because the Llama Stack python client is
not yet updated to accept this text field.

Signed-off-by: Ben Browning <bbrownin@redhat.com>
This commit is contained in:
Ben Browning 2025-06-03 17:43:00 -04:00 • committed by GitHub
parent c70ca8344f
commit 8bee2954be
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
8 changed files with 323 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

86
docs/_static/llama-stack-spec.html vendored
View file

@ -7241,6 +7241,79 @@
],
"title": "OpenAIResponseOutputMessageWebSearchToolCall"
},
"OpenAIResponseText": {
"type": "object",
"properties": {
"format": {
"type": "object",
"properties": {
"type": {
"oneOf": [
{
"type": "string",
"const": "text"
},
{
"type": "string",
"const": "json_schema"
},
{
"type": "string",
"const": "json_object"
}
],
"description": "Must be \"text\", \"json_schema\", or \"json_object\" to identify the format type"
},
"name": {
"type": "string",
"description": "The name of the response format. Only used for json_schema."
},
"schema": {
"type": "object",
"additionalProperties": {
"oneOf": [
{
"type": "null"
},
{
"type": "boolean"
},
{
"type": "number"
},
{
"type": "string"
},
{
"type": "array"
},
{
"type": "object"
}
]
},
"description": "The JSON schema the response should conform to. In a Python SDK, this is often a `pydantic` model. Only used for json_schema."
},
"description": {
"type": "string",
"description": "(Optional) A description of the response format. Only used for json_schema."
},
"strict": {
"type": "boolean",
"description": "(Optional) Whether to strictly enforce the JSON schema. If true, the response must match the schema exactly. Only used for json_schema."
}
},
"additionalProperties": false,
"required": [
"type"
],
"title": "OpenAIResponseTextFormat",
"description": "Configuration for Responses API text format."
}
},
"additionalProperties": false,
"title": "OpenAIResponseText"
},
"CreateOpenaiResponseRequest": {
"type": "object",
"properties": {
@ -7278,6 +7351,9 @@
"temperature": {
"type": "number"
},
"text": {
"$ref": "#/components/schemas/OpenAIResponseText"
},
"tools": {
"type": "array",
"items": {
@ -7351,6 +7427,9 @@
"temperature": {
"type": "number"
},
"text": {
"$ref": "#/components/schemas/OpenAIResponseText"
},
"top_p": {
"type": "number"
},
@ -7369,7 +7448,8 @@
"object",
"output",
"parallel_tool_calls",
"status"
"status",
"text"
],
"title": "OpenAIResponseObject"
},
@ -10406,6 +10486,9 @@
"temperature": {
"type": "number"
},
"text": {
"$ref": "#/components/schemas/OpenAIResponseText"
},
"top_p": {
"type": "number"
},
@ -10431,6 +10514,7 @@
"output",
"parallel_tool_calls",
"status",
"text",
"input"
],
"title": "OpenAIResponseObjectWithInput"