phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-28 03:42:00 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Fixes for multi-turn tool calls in Responses API

Browse source 
Testing with Codex locally, I found another issue in how we were
plumbing through tool calls in multi-turn scenarios and the way tool
call inputs and outputs from previous turns were passed back into
future turns.

This led me to realize we were missing the function tool call output
type in the Responses API, so this adds that and plumbs handling of it
through the responses API to chat completion conversion code.

Signed-off-by: Ben Browning <bbrownin@redhat.com>
This commit is contained in:
Ben Browning 2025-05-08 16:21:15 -04:00
parent 65c56d0ee8
commit 4df8caab41
4 changed files with 187 additions and 69 deletions
Download patch file Download diff file Expand all files Collapse all files

15
llama_stack/apis/agents/openai_responses.py
View file

@ -130,9 +130,24 @@ OpenAIResponseObjectStream = Annotated[
register_schema(OpenAIResponseObjectStream, name="OpenAIResponseObjectStream")
@json_schema_type
class OpenAIResponseInputFunctionToolCallOutput(BaseModel):
"""
This represents the output of a function call that gets passed back to the model.
"""
call_id: str
output: str
type: Literal["function_call_output"] = "function_call_output"
id: str | None = None
status: str | None = None
OpenAIResponseInput = Annotated[
# Responses API allows output messages to be passed in as input
OpenAIResponseOutputMessageWebSearchToolCall
| OpenAIResponseOutputMessageFunctionToolCall
| OpenAIResponseInputFunctionToolCallOutput
|
# Fallback to the generic message type as a last resort
OpenAIResponseMessage,