Skip to content

Google: Continue on MALFORMED_FUNCTION_CALL #3095

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Google: Continue on MALFORMED_FUNCTION_CALL #3095

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions pydantic_ai_slim/pydantic_ai/models/google.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -640,6 +640,8 @@ async def _get_event_iterator(self) -> AsyncIterator[ModelResponseStreamEvent]:
if candidate.finish_reason == 'STOP': # pragma: no cover
# Normal completion - skip this chunk
continue
elif candidate.finish_reason == 'MALFORMED_FUNCTION_CALL': # pragma: no cover
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can we add a test for this, based on a real-world response with this finish reason?

Is skipping the chunk the right thing to do? Would any more chunks follow or would this be the last one in the response? Is it valid to just move on and ignore the error?

If the response before this failure is empty, we retry the request, which I suppose is the goal:

pydantic-ai/pydantic_ai_slim/pydantic_ai/_agent_graph.py

Lines 567 to 572 in 9b1913e

# Go back to the model request node with an empty request, which means we'll essentially
# resubmit the most recent request that resulted in an empty response,
# as the empty response and request will not create any items in the API payload,
# in the hope the model will return a non-empty response this time.
ctx.state.increment_retries(ctx.deps.max_result_retries)
self._next_node = ModelRequestNode[DepsT, NodeRunEndT](_messages.ModelRequest(parts=[]))

continue
elif candidate.finish_reason == 'SAFETY': # pragma: no cover
raise UnexpectedModelBehavior('Safety settings triggered', str(chunk))
else: # pragma: no cover
Expand Down