[Optim][Qwen3TTS] big boost model throughput+latency high concurrency (#1852)

Signed-off-by: pablo <pablo@agigo.ai>
Signed-off-by: JuanPZuluaga <juanz9312@gmail.com>
Co-authored-by: pablo <pablo@agigo.ai>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
This commit is contained in:
authored and GitHub committed 2026-03-18 12:22:05 -04:00
1 parent 53ebdd0b1d
commit fc5e8f825e
5 files changed
+75 -30

No files matched your search

@@ -54,6 +54,8 @@ def build_adapter(monkeypatch, mocker: MockerFixture):
self._pending_save_reqs = deque()
self._finished_save_reqs = set()
self.stop_event = threading.Event()
self._recv_cond = threading.Condition()
self._save_cond = threading.Condition()
monkeypatch.setattr(OmniTransferAdapterBase, "__init__", _fake_base_init)
monkeypatch.setattr(