Skip to content

perf: optimize RunState memory with slots and faster list operations #2867

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

Closed
KirobotDev wants to merge 10 commits into from
+9 −13
Closed

perf: optimize RunState memory with slots and faster list operations #2867

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
3d1473c
perf: optimize RunState memory with slots and faster list operations
KirobotDev Apr 9, 2026
38ecda4
Update run.py
KirobotDev Apr 9, 2026
fa2e7ab
Update run.py
KirobotDev Apr 9, 2026
28fba17
Update run.py
KirobotDev Apr 9, 2026
6dc3b80
Update run_state.py
KirobotDev Apr 9, 2026
39238a7
Update run_state.py
KirobotDev Apr 9, 2026
256ba8e
Update run_state.py
KirobotDev Apr 9, 2026
06cddec
fix: prevent shared mutable state in max_turns handler and simplify l…
KirobotDev Apr 9, 2026
b53873c
Merge branch 'main' of https://github.com/KirobotDev/openai-agents-py…
KirobotDev Apr 9, 2026
829a77e
Merge branch 'main' into main
KirobotDev Apr 10, 2026
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
20 changes: 8 additions & 12 deletions src/agents/run.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -797,12 +797,12 @@ def _with_reasoning_item_id_policy(result: RunResult) -> RunResult:
max_turns=max_turns,
)
result._current_turn = current_turn
result._model_input_items = list(generated_items)
result._model_input_items = generated_items[:]
# Keep normalized replay aligned with the model-facing
# continuation whenever session history preserved extra items.
result._replay_from_model_input_items = list(
generated_items
) != list(session_items)
result._replay_from_model_input_items = (
generated_items != session_items
)
if run_state is not None:
result._trace_state = run_state._trace_state
if session_persistence_enabled:
Expand Down Expand Up @@ -936,10 +936,8 @@ def _with_reasoning_item_id_policy(result: RunResult) -> RunResult:
max_turns=max_turns,
)
result._current_turn = max_turns
result._model_input_items = list(generated_items)
result._replay_from_model_input_items = list(generated_items) != list(
session_items
)
result._model_input_items = generated_items[:]
result._replay_from_model_input_items = generated_items != session_items
if run_state is not None:
result._trace_state = run_state._trace_state
if session_persistence_enabled and include_in_history:
Expand Down Expand Up @@ -1207,10 +1205,8 @@ def _with_reasoning_item_id_policy(result: RunResult) -> RunResult:
max_turns=max_turns,
)
result._current_turn = current_turn
result._model_input_items = list(generated_items)
result._replay_from_model_input_items = list(generated_items) != list(
session_items
)
result._model_input_items = generated_items[:]
result._replay_from_model_input_items = generated_items != session_items
if run_state is not None:
result._current_turn_persisted_item_count = (
run_state._current_turn_persisted_item_count
Expand Down
2 changes: 1 addition & 1 deletion src/agents/run_state.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -135,7 +135,7 @@
_MISSING_CONTEXT_SENTINEL = object()


@dataclass
@dataclass(slots=True)
class RunState(Generic[TContext, TAgent]):
"""Serializable snapshot of an agent run, including context, usage, and interruptions.
Expand Down