diff --git a/backend/app/services/ai_provider.py b/backend/app/services/ai_provider.py
index e7aae8c..1e71bb0 100644
--- a/backend/app/services/ai_provider.py
+++ b/backend/app/services/ai_provider.py
@@ -504,6 +504,7 @@ def _openai_client(api_key: str, timeout: float):
 # 只在 400 明确指出对应参数时移除该参数并重试; 每个参数最多移除一次。
 _TEMP_REJECT_HINTS = ("temperature", "only 1 is allowed")
 _REASONING_EFFORT_REJECT_HINTS = ("reasoning_effort", "reasoning effort")
+_THINKING_BODY_REJECT_HINTS = ("thinking",)
 
 
 def _is_temperature_rejected(exc: Exception) -> bool:
@@ -526,6 +527,16 @@ def _is_reasoning_effort_rejected(exc: Exception) -> bool:
     )
 
 
+def _is_thinking_body_rejected(exc: Exception) -> bool:
+    """True if the upstream 400 specifically rejects the thinking extra_body."""
+    if getattr(exc, "status_code", None) != 400:
+        return False
+    text = _openai_error_detail(exc) or str(exc)
+    return _openai_error_param(exc) == "thinking" or any(
+        h in text.lower() for h in _THINKING_BODY_REJECT_HINTS
+    )
+
+
 def _openai_error_param(exc: Exception) -> str:
     body = getattr(exc, "body", None)
     if not isinstance(body, dict):
@@ -545,6 +556,11 @@ def _openai_retry_kwargs(exc: Exception, kwargs: dict) -> dict | None:
     if "reasoning_effort" in retry_kwargs and _is_reasoning_effort_rejected(exc):
         retry_kwargs.pop("reasoning_effort")
         return retry_kwargs
+    if "extra_body" in retry_kwargs and _is_thinking_body_rejected(exc):
+        # DeepSeek thinking 禁用参数被拒 (模型/API 版本差异): 回退默认思考模式
+        # 重试; 报告若因此被推理挤占正文, 由 _iter_openai_text 显式报错。
+        retry_kwargs.pop("extra_body")
+        return retry_kwargs
     return None
 
 
diff --git a/backend/tests/test_ai_provider.py b/backend/tests/test_ai_provider.py
index 7995e56..98eb8fe 100644
--- a/backend/tests/test_ai_provider.py
+++ b/backend/tests/test_ai_provider.py
@@ -198,6 +198,36 @@ def test_is_temperature_rejected_false_for_other_400():
     assert _is_temperature_rejected(exc) is False
 
 
+def test_thinking_body_rejected_falls_back_to_default_mode():
+    """DeepSeek thinking 禁用参数被 400 拒绝时, 去参重试而非直接失败。"""
+    response = httpx.Response(
+        400,
+        json={"error": {"message": "unknown parameter: thinking"}},
+        request=httpx.Request("POST", "https://api.deepseek.com/v1/chat/completions"),
+    )
+    exc = openai.BadRequestError(
+        "bad request", response=response,
+        body={"error": {"message": "unknown parameter: thinking"}},
+    )
+    kwargs = {"max_tokens": None, "extra_body": {"thinking": {"type": "disabled"}}}
+    retry = ai_provider._openai_retry_kwargs(exc, kwargs)
+    assert retry == {"max_tokens": None}
+    # 原 kwargs 不被修改
+    assert kwargs["extra_body"] == {"thinking": {"type": "disabled"}}
+
+    # 与 thinking 无关的 400 不触发该回退
+    response_other = httpx.Response(
+        400,
+        json={"error": {"message": "model not found"}},
+        request=httpx.Request("POST", "https://api.deepseek.com/v1/chat/completions"),
+    )
+    exc_other = openai.BadRequestError(
+        "bad request", response=response_other,
+        body={"error": {"message": "model not found"}},
+    )
+    assert ai_provider._openai_retry_kwargs(exc_other, kwargs) is None
+
+
 def test_is_temperature_rejected_false_for_non_400():
     response = httpx.Response(
         401,