fix: add reasoning_content field for DeepSeek v4 models in assistant messages

astrbot/core/agent/runners/tool_loop_agent_runner.py

@@ -183,10 +183,10 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
         self.stats.end_time = time.time()
 
         parts = []
-        if llm_resp.reasoning_content or llm_resp.reasoning_signature:
+        if llm_resp.reasoning_content is not None or llm_resp.reasoning_signature:
             parts.append(
                 ThinkPart(
-                    think=llm_resp.reasoning_content,
+                    think=llm_resp.reasoning_content or "",
                     encrypted=llm_resp.reasoning_signature,
                 )
             )
@@ -876,10 +876,10 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
 
             # 将结果添加到上下文中
             parts = []
-            if llm_resp.reasoning_content or llm_resp.reasoning_signature:
+            if llm_resp.reasoning_content is not None or llm_resp.reasoning_signature:
                 parts.append(
                     ThinkPart(
-                        think=llm_resp.reasoning_content,
+                        think=llm_resp.reasoning_content or "",
                         encrypted=llm_resp.reasoning_signature,
                     )
                 )
@@ -1361,10 +1361,10 @@ class ToolLoopAgentRunner(BaseAgentRunner[TContext]):
         self.stats.end_time = time.time()
 
         parts = []
-        if llm_resp.reasoning_content or llm_resp.reasoning_signature:
+        if llm_resp.reasoning_content is not None or llm_resp.reasoning_signature:
             parts.append(
                 ThinkPart(
-                    think=llm_resp.reasoning_content,
+                    think=llm_resp.reasoning_content or "",
                     encrypted=llm_resp.reasoning_signature,
                 )
             )

astrbot/core/provider/entities.py

@@ -353,7 +353,7 @@ class LLMResponse:
     """Tool call IDs."""
     tools_call_extra_content: dict[str, dict[str, Any]] = field(default_factory=dict)
     """Tool call extra content. tool_call_id -> extra_content dict"""
-    reasoning_content: str = ""
+    reasoning_content: str | None = None
     """The reasoning content extracted from the LLM, if any."""
     reasoning_signature: str | None = None
     """The signature of the reasoning content, if any."""
@@ -404,8 +404,6 @@ class LLMResponse:
             raw_completion (ChatCompletion, optional): 原始响应, OpenAI 格式. Defaults to None.
 
         """
-        if reasoning_content is None:
-            reasoning_content = ""
         if tools_call_args is None:
             tools_call_args = []
         if tools_call_name is None:

astrbot/core/provider/sources/anthropic_source.py

@@ -39,7 +39,7 @@ class ProviderAnthropic(Provider):
         stop_reason: str | None = None,
     ) -> None:
         has_text_output = bool((llm_response.completion_text or "").strip())
-        has_reasoning_output = bool(llm_response.reasoning_content.strip())
+        has_reasoning_output = bool((llm_response.reasoning_content or "").strip())
         has_tool_output = bool(llm_response.tools_call_args)
         if has_text_output or has_reasoning_output or has_tool_output:
             return

astrbot/core/provider/sources/gemini_source.py

@@ -462,7 +462,7 @@ class ProviderGoogleGenAI(Provider):
         finish_reason: str | None = None,
     ) -> None:
         has_text_output = bool((llm_response.completion_text or "").strip())
-        has_reasoning_output = bool(llm_response.reasoning_content.strip())
+        has_reasoning_output = bool((llm_response.reasoning_content or "").strip())
         has_tool_output = bool(llm_response.tools_call_args)
         if has_text_output or has_reasoning_output or has_tool_output:
             return

astrbot/core/provider/sources/openai_source.py

@@ -671,9 +671,9 @@ class ProviderOpenAIOfficial(Provider):
             reasoning = self._extract_reasoning_content(chunk)
             _y = False
             llm_response.id = chunk.id
-            llm_response.reasoning_content = ""
+            llm_response.reasoning_content = None
             llm_response.completion_text = ""
-            if reasoning:
+            if reasoning is not None:
                 llm_response.reasoning_content = reasoning
                 _y = True
             if delta and delta.content:
@@ -701,22 +701,28 @@ class ProviderOpenAIOfficial(Provider):
     def _extract_reasoning_content(
         self,
         completion: ChatCompletion | ChatCompletionChunk,
-    ) -> str:
+    ) -> str | None:
         """Extract reasoning content from OpenAI ChatCompletion if available."""
-        reasoning_text = ""
+
+        def _get_reasoning_attr(obj: Any) -> str | None:
+            fields_set = getattr(obj, "model_fields_set", None)
+            if isinstance(fields_set, set) and self.reasoning_key in fields_set:
+                attr = getattr(obj, self.reasoning_key, "")
+                return "" if attr is None else str(attr)
+            attr = getattr(obj, self.reasoning_key, None)
+            return None if attr is None else str(attr)
+
         if not completion.choices:
-            return reasoning_text
+            return None
         if isinstance(completion, ChatCompletion):
             choice = completion.choices[0]
-            reasoning_attr = getattr(choice.message, self.reasoning_key, None)
-            if reasoning_attr:
-                reasoning_text = str(reasoning_attr)
+            reasoning_attr = _get_reasoning_attr(choice.message)
         elif isinstance(completion, ChatCompletionChunk):
             delta = completion.choices[0].delta
-            reasoning_attr = getattr(delta, self.reasoning_key, None)
-            if reasoning_attr:
-                reasoning_text = str(reasoning_attr)
-        return reasoning_text
+            reasoning_attr = _get_reasoning_attr(delta)
+        else:
+            return None
+        return reasoning_attr
 
     def _extract_usage(self, usage: CompletionUsage | dict) -> TokenUsage:
         ptd = getattr(usage, "prompt_tokens_details", None)
@@ -859,7 +865,9 @@ class ProviderOpenAIOfficial(Provider):
 
         # parse the reasoning content if any
         # the priority is higher than the <think> tag extraction
-        llm_response.reasoning_content = self._extract_reasoning_content(completion)
+        reasoning_content = self._extract_reasoning_content(completion)
+        if reasoning_content is not None:
+            llm_response.reasoning_content = reasoning_content
 
         # parse tool calls if any
         if choice.message.tool_calls and tools is not None:
@@ -906,7 +914,7 @@ class ProviderOpenAIOfficial(Provider):
                 "API 返回的 completion 由于内容安全过滤被拒绝(非 AstrBot)。",
             )
         has_text_output = bool((llm_response.completion_text or "").strip())
-        has_reasoning_output = bool(llm_response.reasoning_content.strip())
+        has_reasoning_output = bool((llm_response.reasoning_content or "").strip())
         if (
             not has_text_output
             and not has_reasoning_output
@@ -987,31 +995,34 @@ class ProviderOpenAIOfficial(Provider):
             model in deepseek_reasoning_models
             or "api.deepseek.com" in self.client.base_url.host
         )
-
         for message in payloads.get("messages", []):
             if message.get("role") == "assistant" and isinstance(
                 message.get("content"), list
             ):
                 reasoning_content = ""
+                reasoning_content_present = False
                 new_content = []  # not including think part
                 for part in message["content"]:
                     if part.get("type") == "think":
+                        reasoning_content_present = True
                         reasoning_content += str(part.get("think"))
                     else:
                         new_content.append(part)
                 # Some providers (Grok, etc.) reject empty content lists.
                 # When all parts were think blocks, fall back to None.
                 message["content"] = new_content or None
-                if is_deepseek_v4_reasoning and not reasoning_content:
-                    logger.info(
-                        "Deepseek v4 model requires non-empty reasoning content, but got empty. Setting to 'none' to satisfy the requirement."
-                    )
-                    # Deepseek models require the field on assistant
-                    # history messages, even when the reasoning content is empty.
-                    message["reasoning_content"] = "none"
-                elif reasoning_content:
+                if reasoning_content_present:
                     message["reasoning_content"] = reasoning_content
 
+            if (
+                message.get("role") == "assistant"
+                and is_deepseek_v4_reasoning
+                and "reasoning_content" not in message
+            ):
+                # DeepSeek v4 reasoning models require the field on assistant
+                # history messages, even when the reasoning content is empty.
+                message["reasoning_content"] = ""
+
             # Gemini 的 function_response 要求 google.protobuf.Struct（即 JSON 对象），
             # 纯文本会触发 400 Invalid argument，需要包一层 JSON。
             if is_gemini and message.get("role") == "tool":