Skip to content

fix(qwen): always send explicit enable_thinking in chat_template_kwargs #109

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
LESdylan wants to merge 1 commit into
base: main
Choose a base branch
from
+6 −5
Open

fix(qwen): always send explicit enable_thinking in chat_template_kwargs #109

Show file tree
Hide file tree
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
5 changes: 3 additions & 2 deletions skillopt/model/qwen_backend.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -191,8 +191,9 @@ def _chat_messages_impl(
"messages": _json_safe(messages),
"max_tokens": min(max_completion_tokens, config.max_tokens),
}
if config.enable_thinking:
payload["chat_template_kwargs"] = {"enable_thinking": True}
# Always send an explicit value: omitting it leaves thinking mode to the
# vLLM server / chat-template default, which varies across deployments (#90).
payload["chat_template_kwargs"] = {"enable_thinking": bool(config.enable_thinking)}
if config.temperature is not None:
payload["temperature"] = config.temperature
if tools:
Expand Down
6 changes: 3 additions & 3 deletions tests/test_qwen_backend.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -149,7 +149,7 @@ def _record_urlopen(
return recorder


def test_chat_target_omits_chat_template_kwargs_when_thinking_disabled(
def test_chat_target_sends_explicit_false_when_thinking_disabled(
monkeypatch: pytest.MonkeyPatch,
isolate_qwen_state: tuple[Any, Any],
) -> None:
Expand All @@ -167,7 +167,7 @@ def test_chat_target_omits_chat_template_kwargs_when_thinking_disabled(

assert text == "<answer>yes</answer>"
assert usage["total_tokens"] == 3
assert "chat_template_kwargs" not in recorder.calls[0]["payload"]
assert recorder.calls[0]["payload"]["chat_template_kwargs"] == {"enable_thinking": False}
assert recorder.calls[0]["timeout"] == 10.0


Expand Down Expand Up @@ -224,4 +224,4 @@ def test_configure_qwen_chat_runtime_toggle_controls_payload(
model_module.chat_target("system", "user", max_completion_tokens=128, retries=1)

assert recorder.calls[0]["payload"]["chat_template_kwargs"] == {"enable_thinking": True}
assert "chat_template_kwargs" not in recorder.calls[1]["payload"]
assert recorder.calls[1]["payload"]["chat_template_kwargs"] == {"enable_thinking": False}