Leon/docutranslate
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: thinking mode was never actually disabled (extra_body bug)

Browse Source 
_add_thinking_mode set data["extra_body"]={"enable_thinking": False},
but extra_body is an OpenAI Python SDK convention that gets unwrapped
to top-level by the SDK. This project sends requests via httpx, so
extra_body was sent as a literal nested key that DashScope ignores —
qwen3.6-plus kept thinking ON, wasting 96% of output tokens on
reasoning (measured: 0.54K reasoning / 0.56K output) and inflating
a 3-min translation to 24-min.

Fix: when the field is extra_body (aliyuncs/google providers), merge
its dict into the top-level request body instead of nesting it.
This commit is contained in:
晏永龙
2026-07-27 18:05:53 +08:00
parent 746425528f
commit d252ac5de3
1 changed files with 7 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
docutranslate/agents/agent.py
View File

@@ -570,10 +570,13 @@ class Agent:
if thinking_mode_result is None: if thinking_mode_result is None:
return return
field_thinking, val_enable, val_disable = thinking_mode_result field_thinking, val_enable, val_disable = thinking_mode_result
if self.thinking == "enable": value = val_enable if self.thinking == "enable" else val_disable
data[field_thinking] = val_enable # extra_body 是 OpenAI SDK 约定；本项目用 httpx 直接发请求，
elif self.thinking == "disable": # 必须把里面的字段提到顶层，DashScope / GLM 等才会识别。
data[field_thinking] = val_disable if field_thinking == "extra_body" and isinstance(value, dict):
data.update(value)
else:
data[field_thinking] = value
def _normalize_mt_lang(self, lang: str | None) -> str | None: def _normalize_mt_lang(self, lang: str | None) -> str | None:
if lang is None: if lang is None: