mergin main，速度提升一下

main/xiaozhi-server/core/providers/llm/dify/dify.py

@@ -10,11 +10,13 @@ class LLMProvider(LLMProviderBase):
     def __init__(self, config):
         self.api_key = config["api_key"]
         self.base_url = config.get("base_url", "https://api.dify.ai/v1").rstrip('/')
+        self.session_conversation_map = {}  # 存储session_id和conversation_id的映射
 
     def response(self, session_id, dialogue):
         try:
             # 取最后一条用户消息
             last_msg = next(m for m in reversed(dialogue) if m["role"] == "user")
+            conversation_id = self.session_conversation_map.get(session_id)
 
             # 发起流式请求
             with requests.post(
@@ -24,13 +26,18 @@ class LLMProvider(LLMProviderBase):
                         "query": last_msg["content"],
                         "response_mode": "streaming",
                         "user": session_id,
-                        "inputs": {}
+                        "inputs": {},
+                        "conversation_id": conversation_id
                     },
                     stream=True
             ) as r:
                 for line in r.iter_lines():
                     if line.startswith(b'data: '):
                         event = json.loads(line[6:])
+                        # 如果没有找到conversation_id，则获取此次conversation_id
+                        if not conversation_id:
+                            conversation_id = event.get('conversation_id')
+                            self.session_conversation_map[session_id] = conversation_id  # 更新映射
                         if event.get('answer'):
                             yield event['answer']
 

main/xiaozhi-server/core/providers/llm/ollama/ollama.py

@@ -12,10 +12,10 @@ class LLMProvider(LLMProviderBase):
         self.model_name = config.get("model_name")
         self.base_url = config.get("base_url", "http://localhost:11434")
         # Initialize OpenAI client with Ollama base URL
-        #如果没有v1，增加v1
+        # 如果没有v1，增加v1
         if not self.base_url.endswith("/v1"):
             self.base_url = f"{self.base_url}/v1"
-            
+
         self.client = OpenAI(
             base_url=self.base_url,
             api_key="ollama"  # Ollama doesn't need an API key but OpenAI client requires one
@@ -28,13 +28,20 @@ class LLMProvider(LLMProviderBase):
                 messages=dialogue,
                 stream=True
             )
-            
+            is_active=True
             for chunk in responses:
                 try:
                     delta = chunk.choices[0].delta if getattr(chunk, 'choices', None) else None
                     content = delta.content if hasattr(delta, 'content') else ''
                     if content:
-                        yield content
+                        if '<think>' in content:
+                            is_active = False
+                            content = content.split('<think>')[0]
+                        if '</think>' in content:
+                            is_active = True
+                            content = content.split('</think>')[-1]
+                        if is_active:
+                            yield content
                 except Exception as e:
                     logger.bind(tag=TAG).error(f"Error processing chunk: {e}")
 
@@ -50,10 +57,10 @@ class LLMProvider(LLMProviderBase):
                 stream=True,
                 tools=functions,
             )
-            
+
             for chunk in stream:
                 yield chunk.choices[0].delta.content, chunk.choices[0].delta.tool_calls
 
         except Exception as e:
             logger.bind(tag=TAG).error(f"Error in Ollama function call: {e}")
-            yield {"type": "content", "content": f"【Ollama服务响应异常: {str(e)}】"}
+            yield {"type": "content", "content": f"【Ollama服务响应异常: {str(e)}】"}