[Optimization] Improve perf for fd response token with internal adapter (#4992)

* [Optimize] Improve perf for fd response token with internal adapter

* fix

* fix bug

* fix ci

* fix ci

* fix ci

* fix ci
This commit is contained in:
chenjian
2025-11-21 19:02:03 +08:00
committed by GitHub
parent 5bcf79d780
commit 3ea1b44a58
15 changed files with 202 additions and 67 deletions

View File

@@ -436,6 +436,5 @@ class SplitwiseConnector:
self.logger.debug(f"_handle_decode function receive {payload}")
tasks = []
for task in payload:
output = RequestOutput.from_dict(task)
tasks.append(output)
tasks.append(RequestOutput.from_dict(task))
self.engine_worker_queue.put_disaggregated_tasks(("decode", tasks))