[Perf] Optimize requests abort (#32211)

Signed-off-by: yewentao256 <zhyanwentao@126.com>

[Perf] Optimize requests abort (#32211)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
2a719e08 · Wentao Ye · GitHub · f243abc9 · 2a719e08
Unverified Commit 2a719e08 authored Jan 12, 2026 by Wentao Ye Committed by GitHub Jan 13, 2026
Show whitespace changes
Inline Side-by-side

Showing with 4 additions and 3 deletions

vllm/v1/engine/async_llm.py vllm/v1/engine/async_llm.py +4 -3

No files found.
--- a/vllm/v1/engine/async_llm.py
+++ b/vllm/v1/engine/async_llm.py
@@ -525,6 +525,7 @@ class AsyncLLM(EngineClient):
                            await asyncio.sleep(0)

                        # 3) Abort any reqs that finished due to stop strings.
+                        if processed_outputs.reqs_to_abort:
                            await engine_core.abort_requests_async(
                                processed_outputs.reqs_to_abort
                            )