update pay function

mcp_server.py

@@ -1999,7 +1999,7 @@ class MCPAgentIntegrated:
             model=self.kimi_model,
             messages=messages,
             temperature=1.0,  # Kimi 推荐
-            max_tokens=16000,  # 足够容纳 reasoning_content
+            max_tokens=8192,  # 足够容纳 reasoning_content
         )
 
         choice = response.choices[0]
@@ -2074,7 +2074,7 @@ class MCPAgentIntegrated:
                 model=self.deepmoney_model,
                 messages=messages,
                 temperature=0.7,
-                max_tokens=1000,
+                max_tokens=8192,
             )
 
             summary = response.choices[0].message.content
@@ -2241,7 +2241,7 @@ class MCPAgentIntegrated:
                 model="kimi-k2-turbo-preview",  # 使用非思考模型，更快
                 messages=messages,
                 temperature=0.7,
-                max_tokens=2000,  # 增加 token 限制以支持图表配置
+                max_tokens=8192,  # 增加 token 限制以支持图表配置
             )
 
             summary = response.choices[0].message.content
@@ -2328,7 +2328,7 @@ class MCPAgentIntegrated:
                 model=self.deepmoney_model,
                 messages=messages,
                 temperature=0.3,
-                max_tokens=100,
+                max_tokens=4096,
             )
 
             title = response.choices[0].message.content.strip()
@@ -2423,7 +2423,7 @@ class MCPAgentIntegrated:
                     model=planning_model,
                     messages=messages,
                     temperature=1.0,
-                    max_tokens=16000,
+                    max_tokens=8192,
                     stream=True,  # 启用流式输出
                 )
 
@@ -2467,7 +2467,7 @@ class MCPAgentIntegrated:
                             model=self.deepmoney_model,
                             messages=messages,
                             temperature=0.7,
-                            max_tokens=16000,
+                            max_tokens=8192,
                         )
 
                         plan_content = fallback_response.choices[0].message.content
@@ -2663,7 +2663,7 @@ class MCPAgentIntegrated:
                         model="kimi-k2-turbo-preview",
                         messages=messages,
                         temperature=0.7,
-                        max_tokens=2000,
+                        max_tokens=8192,
                         stream=True,  # 启用流式输出
                     )
 
@@ -2697,7 +2697,7 @@ class MCPAgentIntegrated:
                                 model=self.deepmoney_model,
                                 messages=messages,
                                 temperature=0.7,
-                                max_tokens=2000,
+                                max_tokens=8192,
                             )
 
                             final_summary = fallback_response.choices[0].message.content
@@ -3709,7 +3709,7 @@ async def stream_role_response(
             messages=messages,
             stream=True,
             temperature=0.7,
-            max_tokens=16384,  # 大幅增加 token 限制以避免输出被截断
+            max_tokens=8192,  # 大幅增加 token 限制以避免输出被截断
         )
 
         full_content = ""