diff --git a/src/chat/chat_loop/cycle_processor.py b/src/chat/chat_loop/cycle_processor.py
index 396f88387..3d8f44dc0 100644
--- a/src/chat/chat_loop/cycle_processor.py
+++ b/src/chat/chat_loop/cycle_processor.py
@@ -1,6 +1,8 @@
 import asyncio
 import time
 import traceback
+import math
+import random
 from typing import Optional, Dict, Any, Tuple
 
 from src.chat.message_receive.chat_stream import get_chat_manager
@@ -10,7 +12,7 @@ from src.config.config import global_config
 from src.chat.planner_actions.planner import ActionPlanner
 from src.chat.planner_actions.action_modifier import ActionModifier
 from src.person_info.person_info import get_person_info_manager
-from src.plugin_system.apis import database_api
+from src.plugin_system.apis import database_api, generator_api
 from src.plugin_system.base.component_types import ChatMode
 from src.mais4u.constant_s4u import ENABLE_S4U
 from src.chat.chat_loop.hfc_utils import send_typing, stop_typing
@@ -44,7 +46,6 @@ class CycleProcessor:
     async def _send_and_store_reply(
         self,
         response_set,
-        reply_to_str,
         loop_start_time,
         action_message,
         cycle_timers: Dict[str, float],
@@ -52,7 +53,7 @@ class CycleProcessor:
         actions,
     ) -> Tuple[Dict[str, Any], str, Dict[str, float]]:
         with Timer("回复发送", cycle_timers):
-            reply_text = await self.response_handler.send_response(response_set, reply_to_str, loop_start_time, action_message)
+            reply_text = await self.response_handler.send_response(response_set, loop_start_time, action_message)
 
             # 存储reply action信息
         person_info_manager = get_person_info_manager()
@@ -60,7 +61,7 @@ class CycleProcessor:
         # 获取 platform，如果不存在则从 chat_stream 获取，如果还是 None 则使用默认值
         platform = action_message.get("chat_info_platform")
         if platform is None:
-            platform = getattr(self.chat_stream, "platform", "unknown")
+            platform = getattr(self.context.chat_stream, "platform", "unknown")
         
         person_id = person_info_manager.get_person_id(
             platform,
@@ -75,7 +76,7 @@ class CycleProcessor:
             action_prompt_display=action_prompt_display,
             action_done=True,
             thinking_id=thinking_id,
-            action_data={"reply_text": reply_text, "reply_to": reply_to_str},
+            action_data={"reply_text": reply_text},
             action_name="reply",
         )
 
@@ -113,12 +114,6 @@ class CycleProcessor:
         """
         action_type = "no_action"
         reply_text = ""  # 初始化reply_text变量，避免UnboundLocalError
-        reply_to_str = ""  # 初始化reply_to_str变量
-
-        # 根据interest_value计算概率，决定使用哪种planner模式
-        # interest_value越高，越倾向于使用Normal模式
-        import random
-        import math
         
         # 使用sigmoid函数将interest_value转换为概率
         # 当interest_value为0时，概率接近0（使用Focus模式）
@@ -224,40 +219,23 @@ class CycleProcessor:
                         "command": command
                     }
                 else:
-                    # 执行回复动作
                     try:
-                        reply_to_str = await self._build_reply_to_str(action_info["action_message"])
-                    except UserWarning:
-                        logger.warning("选取了自己作为回复对象，跳过回复生成")
-                        return {
-                            "action_type": "reply",
-                            "success": False,
-                            "reply_text": "",
-                            "loop_info": None
-                        }
-                    
-                    # 生成回复
-                    gather_timeout = global_config.chat.thinking_timeout
-                    try:
-                        response_set = await asyncio.wait_for(
-                            self.response_handler.generate_response(
-                                message_data=action_info["action_message"],
-                                available_actions=action_info["available_actions"],
-                                reply_to=reply_to_str,
-                                request_type="chat.replyer",
-                            ),
-                            timeout=gather_timeout
-                        )
-                    except asyncio.TimeoutError:
-                        logger.warning(
-                            f"{self.log_prefix} 并行执行：回复生成超时>{global_config.chat.thinking_timeout}s，已跳过"
-                        )
-                        return {
-                            "action_type": "reply",
-                            "success": False,
-                            "reply_text": "",
-                            "loop_info": None
-                        }
+                        success, response_set, _ = await generator_api.generate_reply(
+                            chat_stream=self.context.chat_stream,
+                            reply_message = action_info["action_message"],
+                            available_actions=available_actions,
+                            enable_tool=global_config.tool.enable_tool,
+                            request_type="chat.replyer",
+                            from_plugin=False,
+                        )     
+                        if not success or not response_set:
+                            logger.info(f"对 {action_info['action_message'].get('processed_plain_text')} 的回复生成失败")
+                            return {
+                                "action_type": "reply",
+                                "success": False,
+                                "reply_text": "",
+                                "loop_info": None
+                            }
                     except asyncio.CancelledError:
                         logger.debug(f"{self.log_prefix} 并行执行：回复生成任务已被取消")
                         return {
@@ -267,18 +245,8 @@ class CycleProcessor:
                             "loop_info": None
                         }
 
-                    if not response_set:
-                        logger.warning(f"{self.log_prefix} 模型超时或生成回复内容为空")
-                        return {
-                            "action_type": "reply",
-                            "success": False,
-                            "reply_text": "",
-                            "loop_info": None
-                        }
-
                     loop_info, reply_text, cycle_timers_reply = await self._send_and_store_reply(
                         response_set,
-                        reply_to_str,
                         loop_start_time,
                         action_info["action_message"],
                         cycle_timers,
@@ -303,7 +271,6 @@ class CycleProcessor:
                 }
             
         # 创建所有动作的后台任务
-
         action_tasks = [asyncio.create_task(execute_action(action)) for action in actions]
 
         # 并行执行所有任务
@@ -367,7 +334,6 @@ class CycleProcessor:
         self.context.chat_instance.cycle_tracker.end_cycle(loop_info, cycle_timers)
         self.context.chat_instance.cycle_tracker.print_cycle_info(cycle_timers)
 
-        # await self.willing_manager.after_generate_reply_handle(message_data.get("message_id", ""))
         action_type = actions[0]["action_type"] if actions else "no_action"
         # 管理no_reply计数器：当执行了非no_reply动作时，重置计数器
         if action_type != "no_reply":
@@ -688,36 +654,6 @@ class CycleProcessor:
             "action_prompt": "",
         }
 
-    async def _build_reply_to_str(self, message_data: dict):
-        """
-        构建回复目标字符串
-
-        Args:
-            message_data: 消息数据字典
-
-        Returns:
-            str: 格式化的回复目标字符串，格式为"用户名:消息内容"
-
-        功能说明:
-        - 从消息数据中提取平台和用户ID信息
-        - 通过人员信息管理器获取用户昵称
-        - 构建用于回复显示的格式化字符串
-        """
-        from src.person_info.person_info import get_person_info_manager
-
-        person_info_manager = get_person_info_manager()
-        platform = (
-            message_data.get("chat_info_platform")
-            or message_data.get("user_platform")
-            or (self.context.chat_stream.platform if self.context.chat_stream else "default")
-        )
-        user_id = message_data.get("user_id", "")
-        if user_id == str(global_config.bot.qq_account) and platform == global_config.bot.platform:
-            raise UserWarning
-        person_id = person_info_manager.get_person_id(platform, user_id)
-        person_name = await person_info_manager.get_value(person_id, "person_name")
-        return f"{person_name}:{message_data.get('processed_plain_text')}"
-
     def _build_final_loop_info(self, reply_loop_info, action_success, action_reply_text, action_command, plan_result):
         """
         构建最终的循环信息
diff --git a/src/chat/chat_loop/energy_manager.py b/src/chat/chat_loop/energy_manager.py
index d4db96ef5..9358664a0 100644
--- a/src/chat/chat_loop/energy_manager.py
+++ b/src/chat/chat_loop/energy_manager.py
@@ -57,26 +57,6 @@ class EnergyManager:
             await asyncio.sleep(0)
             logger.info(f"{self.context.log_prefix} 能量管理器已停止")
 
-    def _handle_energy_completion(self, task: asyncio.Task):
-        """
-        处理能量循环任务完成
-
-        Args:
-            task: 完成的异步任务对象
-
-        功能说明:
-        - 处理任务正常完成或异常情况
-        - 记录相应的日志信息
-        - 区分取消和异常终止的情况
-        """
-        try:
-            if exception := task.exception():
-                logger.error(f"{self.context.log_prefix} 能量循环异常: {exception}")
-            else:
-                logger.info(f"{self.context.log_prefix} 能量循环正常结束")
-        except asyncio.CancelledError:
-            logger.info(f"{self.context.log_prefix} 能量循环被取消")
-
     async def _energy_loop(self):
         """
         能量与睡眠压力管理的主循环
diff --git a/src/chat/chat_loop/heartFC_chat.py b/src/chat/chat_loop/heartFC_chat.py
index 895fb4b0d..b8922d092 100644
--- a/src/chat/chat_loop/heartFC_chat.py
+++ b/src/chat/chat_loop/heartFC_chat.py
@@ -12,14 +12,12 @@ from src.chat.express.expression_learner import expression_learner_manager
 from src.plugin_system.base.component_types import ChatMode
 from src.schedule.schedule_manager import schedule_manager, SleepState
 from src.plugin_system.apis import message_api
-from src.chat.willing.willing_manager import get_willing_manager
 
 from .hfc_context import HfcContext
 from .energy_manager import EnergyManager
 from .proactive_thinker import ProactiveThinker
 from .cycle_processor import CycleProcessor
 from .response_handler import ResponseHandler
-from .normal_mode_handler import NormalModeHandler
 from .cycle_tracker import CycleTracker
 from .wakeup_manager import WakeUpManager
 
@@ -47,7 +45,6 @@ class HeartFChatting:
         self.cycle_processor = CycleProcessor(self.context, self.response_handler, self.cycle_tracker)
         self.energy_manager = EnergyManager(self.context)
         self.proactive_thinker = ProactiveThinker(self.context, self.cycle_processor)
-        self.normal_mode_handler = NormalModeHandler(self.context, self.cycle_processor)
         self.wakeup_manager = WakeUpManager(self.context)
 
         # 将唤醒度管理器设置到上下文中
@@ -60,7 +57,6 @@ class HeartFChatting:
         
         # 记录最近3次的兴趣度
         self.recent_interest_records: deque = deque(maxlen=3)
-        self.willing_manager = get_willing_manager()
         self._initialize_chat_mode()
         logger.info(f"{self.context.log_prefix} HeartFChatting 初始化完成")
 
@@ -97,7 +93,7 @@ class HeartFChatting:
         self.context.relationship_builder = relationship_builder_manager.get_or_create_builder(self.context.stream_id)
         self.context.expression_learner = expression_learner_manager.get_expression_learner(self.context.stream_id)
 
-        await self.energy_manager.start()
+        #await self.energy_manager.start()
         await self.proactive_thinker.start()
         await self.wakeup_manager.start()
 
@@ -120,7 +116,7 @@ class HeartFChatting:
             return
         self.context.running = False
 
-        await self.energy_manager.stop()
+        #await self.energy_manager.stop()
         await self.proactive_thinker.stop()
         await self.wakeup_manager.stop()
 
@@ -245,8 +241,6 @@ class HeartFChatting:
             # 统一使用 _should_process_messages 判断是否应该处理
             should_process,interest_value = await self._should_process_messages(recent_messages if has_new_messages else None)
             if should_process:
-                #earliest_message_data = recent_messages[0]
-                #self.last_read_time = earliest_message_data.get("time")
                 self.context.last_read_time = time.time()
                 await self.cycle_processor.observe(interest_value = interest_value)
             else:
@@ -418,6 +412,7 @@ class HeartFChatting:
         # talk_frequency = global_config.chat.get_current_talk_frequency(self.context.chat_stream.stream_id)
         modified_exit_count_threshold = self.context.focus_energy / global_config.chat.focus_value
 
+        modified_exit_interest_threshold = 3 / global_config.chat.focus_value
         total_interest = 0.0
         for msg_dict in new_message:
             interest_value = msg_dict.get("interest_value", 0.0)
@@ -441,11 +436,11 @@ class HeartFChatting:
             if not hasattr(self, "_last_accumulated_interest") or total_interest != self._last_accumulated_interest:
                 logger.info(f"{self.context.log_prefix} breaking形式当前累计兴趣值: {total_interest:.2f}, 专注度: {global_config.chat.focus_value:.1f}")
                 self._last_accumulated_interest = total_interest
-            if total_interest >= 3 / global_config.chat.focus_value:
+            if total_interest >= modified_exit_interest_threshold:
                 # 记录兴趣度到列表      
                 self.recent_interest_records.append(total_interest)
                 logger.info(
-                    f"{self.context.log_prefix} 累计兴趣值达到{total_interest:.2f}(>{3 / global_config.chat.focus_value})，结束等待"
+                    f"{self.context.log_prefix} 累计兴趣值达到{total_interest:.2f}(>{modified_exit_interest_threshold:.1f})，结束等待"
                 )
                 return True,total_interest/new_message_count
             
diff --git a/src/chat/chat_loop/normal_mode_handler.py b/src/chat/chat_loop/normal_mode_handler.py
deleted file mode 100644
index a554ab3c4..000000000
--- a/src/chat/chat_loop/normal_mode_handler.py
+++ /dev/null
@@ -1,84 +0,0 @@
-import random
-from typing import Dict, Any, TYPE_CHECKING
-
-from src.common.logger import get_logger
-from src.config.config import global_config
-from src.chat.willing.willing_manager import get_willing_manager
-from .hfc_context import HfcContext
-
-if TYPE_CHECKING:
-    from .cycle_processor import CycleProcessor
-
-logger = get_logger("hfc.normal_mode")
-
-
-class NormalModeHandler:
-    def __init__(self, context: HfcContext, cycle_processor: "CycleProcessor"):
-        """
-        初始化普通模式处理器
-
-        Args:
-            context: HFC聊天上下文对象
-            cycle_processor: 循环处理器，用于处理决定回复的消息
-
-        功能说明:
-        - 处理NORMAL模式下的消息
-        - 根据兴趣度和回复概率决定是否回复
-        - 管理意愿系统和回复概率计算
-        """
-        self.context = context
-        self.cycle_processor = cycle_processor
-        self.willing_manager = get_willing_manager()
-
-    async def handle_message(self, message_data: Dict[str, Any]) -> bool:
-        """
-        处理NORMAL模式下的单条消息
-
-        Args:
-            message_data: 消息数据字典，包含用户信息、消息内容、兴趣值等
-
-        Returns:
-            bool: 是否进行了回复处理
-
-        功能说明:
-        - 计算消息的兴趣度和基础回复概率
-        - 应用谈话频率调整回复概率
-        - 过滤表情和图片消息（设置回复概率为0）
-        - 根据概率随机决定是否回复
-        - 如果决定回复则调用循环处理器进行处理
-        - 记录详细的决策日志
-        """
-        if not self.context.chat_stream:
-            return False
-
-        interested_rate = message_data.get("interest_value") or 0.0
-        self.willing_manager.setup(message_data, self.context.chat_stream)
-        reply_probability = await self.willing_manager.get_reply_probability(message_data.get("message_id", ""))
-
-        if reply_probability < 1:
-            additional_config = message_data.get("additional_config", {})
-            if additional_config and "maimcore_reply_probability_gain" in additional_config:
-                reply_probability += additional_config["maimcore_reply_probability_gain"]
-                reply_probability = min(max(reply_probability, 0), 1)
-
-        talk_frequency = global_config.chat.get_current_talk_frequency(self.context.stream_id)
-        reply_probability = talk_frequency * reply_probability
-
-        if message_data.get("is_emoji") or message_data.get("is_picid"):
-            reply_probability = 0
-
-        mes_name = self.context.chat_stream.group_info.group_name if self.context.chat_stream.group_info else "私聊"
-        if reply_probability > 0.05:
-            logger.info(
-                f"[{mes_name}]"
-                f"{message_data.get('user_nickname')}:"
-                f"{message_data.get('processed_plain_text')}[兴趣:{interested_rate:.2f}][回复概率:{reply_probability * 100:.1f}%]"
-            )
-
-        if random.random() < reply_probability:
-            await self.willing_manager.before_generate_reply_handle(message_data.get("message_id", ""))
-            await self.cycle_processor.observe(message_data=message_data)
-            return True
-
-        self.willing_manager.delete(message_data.get("message_id", ""))
-        return False
diff --git a/src/chat/chat_loop/response_handler.py b/src/chat/chat_loop/response_handler.py
index 3a5ee93e0..ecfc6addb 100644
--- a/src/chat/chat_loop/response_handler.py
+++ b/src/chat/chat_loop/response_handler.py
@@ -105,7 +105,7 @@ class ResponseHandler:
 
         return loop_info, reply_text, cycle_timers
 
-    async def send_response(self, reply_set, reply_to, thinking_start_time, message_data) -> str:
+    async def send_response(self, reply_set, thinking_start_time, message_data) -> str:
         """
         发送回复内容的具体实现
 
@@ -129,9 +129,6 @@ class ResponseHandler:
         new_message_count = message_api.count_new_messages(
             chat_id=self.context.stream_id, start_time=thinking_start_time, end_time=current_time
         )
-        platform = message_data.get("user_platform", "")
-        user_id = message_data.get("user_id", "")
-        reply_to_platform_id = f"{platform}:{user_id}"
 
         need_reply = new_message_count >= random.randint(2, 4)
 
@@ -157,32 +154,26 @@ class ResponseHandler:
                 continue
 
             if not first_replied:
-                if need_reply:
-                    await send_api.text_to_stream(
-                        text=data,
-                        stream_id=self.context.stream_id,
-                        reply_to=reply_to,
-                        reply_to_platform_id=reply_to_platform_id,
-                        typing=False,
-                    )
-                else:
-                    await send_api.text_to_stream(
-                        text=data,
-                        stream_id=self.context.stream_id,
-                        reply_to_platform_id=reply_to_platform_id,
-                        typing=False,
-                    )
+                await send_api.text_to_stream(
+                    text=data,
+                    stream_id=self.context.stream_id,
+                    reply_to_message = message_data,
+                    set_reply=need_reply,
+                    typing=False,
+                )
                 first_replied = True
             else:
                 await send_api.text_to_stream(
                     text=data,
                     stream_id=self.context.stream_id,
-                    reply_to_platform_id=reply_to_platform_id,
+                    reply_to_message = message_data,
+                    set_reply=need_reply,
                     typing=True,
                 )
 
         return reply_text
 
+    # TODO: 已废弃
     async def generate_response(
         self,
         message_data: dict,
diff --git a/src/chat/express/expression_learner.py b/src/chat/express/expression_learner.py
index 3f4f2b291..2f8943c2a 100644
--- a/src/chat/express/expression_learner.py
+++ b/src/chat/express/expression_learner.py
@@ -61,7 +61,7 @@ def init_prompt() -> None:
 class ExpressionLearner:
     def __init__(self, chat_id: str) -> None:
         self.express_learn_model: LLMRequest = LLMRequest(
-            model_set=model_config.model_task_config.replyer, request_type="expression.learner"
+            model_set=model_config.model_task_config.replyer, request_type="expressor.learner"
         )
         self.chat_id = chat_id
         self.chat_name = get_chat_manager().get_stream_name(chat_id) or chat_id
diff --git a/src/chat/planner_actions/planner.py b/src/chat/planner_actions/planner.py
index e09818527..2d945cd94 100644
--- a/src/chat/planner_actions/planner.py
+++ b/src/chat/planner_actions/planner.py
@@ -51,13 +51,13 @@ def init_prompt():
 
 动作：reply
 动作描述：参与聊天回复，发送文本进行表达
-- 你想要闲聊或者随便附和
-- 有人提到了你，但是你还没有回应
+- 你想要闲聊或者随便附
 - {mentioned_bonus}
 - 如果你刚刚进行了回复，不要对同一个话题重复回应
+- 不要回复自己发送的消息
 {{
     "action": "reply",
-    "target_message_id":"想要回复的消息id",
+    "target_message_id":"触发action的消息id",
     "reason":"回复的原因"
 }}
 
@@ -316,7 +316,7 @@ class ActionPlanner:
                         if key not in ["action", "reason"]:
                             action_data[key] = value
 
-                    # 在FOCUS模式下，非no_reply动作需要target_message_id
+                    # 非no_reply动作需要target_message_id
                     if action != "no_reply":
                         if target_message_id := parsed_json.get("target_message_id"):
                             # 根据target_message_id查找原始消息
@@ -341,9 +341,7 @@ class ActionPlanner:
                             logger.warning(f"{self.log_prefix}动作'{action}'缺少target_message_id")
                     
                     
-                    if action == "no_action":
-                        reasoning = "normal决定不使用额外动作"
-                    elif action != "no_reply" and action != "reply" and action not in current_available_actions:
+                    if action != "no_reply" and action != "reply" and action not in current_available_actions:
                         logger.warning(
                             f"{self.log_prefix}LLM 返回了当前不可用或无效的动作: '{action}' (可用: {list(current_available_actions.keys())})，将强制使用 'no_reply'"
                         )
@@ -483,32 +481,15 @@ class ActionPlanner:
                 mentioned_bonus = "\n- 有人提到你，或者at你"
 
             if mode == ChatMode.FOCUS:
-                no_action_block = f"""重要说明：
-- 'no_reply' 表示只进行不进行回复，等待合适的回复时机
+                no_action_block = """重要说明：
+- 'no_reply' 表示不进行回复，等待合适的回复时机
 - 当你刚刚发送了消息，没有人回复时，选择no_reply
 - 当你一次发送了太多消息，为了避免打扰聊天节奏，选择no_reply
-
-动作：reply
-动作描述：参与聊天回复，发送文本进行表达
-- 你想要闲聊或者随便附和{mentioned_bonus}
-- 如果你刚刚进行了回复，不要对同一个话题重复回应
-- 不要回复自己发送的消息
-{{
-    "action": "no_reply",
-    "reason":"不回复的原因"
-}}
 """
             else:  # NORMAL Mode
-                no_action_block = f"""重要说明：
+                no_action_block = """重要说明：
 - 'reply' 表示只进行普通聊天回复，不执行任何额外动作
 - 其他action表示在普通回复的基础上，执行相应的额外动作
-
-动作：reply
-动作描述：参与聊天回复，发送文本进行表达
-- 你想要闲聊或者随便附
-- {mentioned_bonus}
-- 如果你刚刚进行了回复，不要对同一个话题重复回应
-- 不要回复自己发送的消息
 {{
     "action": "reply",
     "target_message_id":"触发action的消息id",
@@ -516,6 +497,7 @@ class ActionPlanner:
 }}"""
 
             chat_context_description = "你现在正在一个群聊中"
+            chat_target_name = None 
             if not is_group_chat and chat_target_info:
                 chat_target_name = (
                     chat_target_info.get("person_name")
@@ -546,6 +528,7 @@ class ActionPlanner:
                 chat_context_description=chat_context_description,
                 chat_content_block=chat_content_block,
                 actions_before_now_block=actions_before_now_block,
+                mentioned_bonus=mentioned_bonus,
                 no_action_block=no_action_block,
                 mentioned_bonus=mentioned_bonus,
                 action_options_text=action_options_block,
diff --git a/src/chat/replyer/default_generator.py b/src/chat/replyer/default_generator.py
index 035fb2422..cb11c0aae 100644
--- a/src/chat/replyer/default_generator.py
+++ b/src/chat/replyer/default_generator.py
@@ -240,14 +240,6 @@ class DefaultReplyer:
 
         self.tool_executor = ToolExecutor(chat_id=self.chat_stream.stream_id)
 
-    def _select_weighted_models_config(self) -> Tuple[TaskConfig, float]:
-        """使用加权随机选择来挑选一个模型配置"""
-        configs = self.model_set
-        # 提取权重，如果模型配置中没有'weight'键，则默认为1.0
-        weights = [weight for _, weight in configs]
-
-        return random.choices(population=configs, weights=weights, k=1)[0]
-
     async def generate_reply_with_context(
         self,
         extra_info: str = "",
@@ -258,7 +250,7 @@ class DefaultReplyer:
         from_plugin: bool = True,
         stream_id: Optional[str] = None,
         reply_message: Optional[Dict[str, Any]] = None,
-    ) -> Tuple[bool, Optional[Dict[str, Any]], Optional[str], List[Dict[str, Any]]]:
+    ) -> Tuple[bool, Optional[Dict[str, Any]], Optional[str]]:
         # sourcery skip: merge-nested-ifs
         """
         回复器 (Replier): 负责生成回复文本的核心逻辑。
@@ -290,7 +282,6 @@ class DefaultReplyer:
                     choosen_actions=choosen_actions,
                     enable_tool=enable_tool,
                     reply_message=reply_message,
-                    reply_reason=reply_reason,
                 )
 
             if not prompt:
@@ -579,7 +570,6 @@ class DefaultReplyer:
         if not enable_tool:
             return ""
 
-
         try:
             # 使用工具执行器获取信息
             tool_results, _, _ = await self.tool_executor.execute_from_chat_message(
@@ -826,7 +816,7 @@ class DefaultReplyer:
         choosen_actions: Optional[List[Dict[str, Any]]] = None,
         enable_tool: bool = True,
         reply_message: Optional[Dict[str, Any]] = None,
-    ) -> Tuple[str, List[int]]:
+    ) -> str:
         """
         构建回复器上下文
 
@@ -838,6 +828,7 @@ class DefaultReplyer:
             enable_timeout: 是否启用超时处理
             enable_tool: 是否启用工具调用
             reply_message: 回复的原始消息
+
         Returns:
             str: 构建好的上下文
         """
@@ -870,7 +861,29 @@ class DefaultReplyer:
                 mood_prompt = f"{mood_prompt}。{angry_prompt_addition}"
         else:
             mood_prompt = ""
-            
+
+        if reply_to:
+            #兼容旧的reply_to
+            sender, target = self._parse_reply_target(reply_to)
+        else:
+            # 获取 platform，如果不存在则从 chat_stream 获取，如果还是 None 则使用默认值
+            platform = reply_message.get("chat_info_platform")
+            person_id = person_info_manager.get_person_id(
+                platform,  # type: ignore
+                reply_message.get("user_id"),  # type: ignore
+            )
+            person_name = await person_info_manager.get_value(person_id, "person_name")
+            sender = person_name
+            target = reply_message.get('processed_plain_text')
+
+        person_info_manager = get_person_info_manager()
+        person_id = person_info_manager.get_person_id_by_person_name(sender)
+        user_id = person_info_manager.get_value_sync(person_id, "user_id")
+        platform = chat_stream.platform
+        if user_id == global_config.bot.qq_account and platform == global_config.bot.platform:
+            logger.warning("选取了自身作为回复对象，跳过构建prompt")
+            return ""
+
         target = replace_user_references_sync(target, chat_stream.platform, replace_bot_name=True)
 
 
@@ -905,11 +918,11 @@ class DefaultReplyer:
                 self.build_expression_habits(chat_talking_prompt_short, target), "expression_habits"
             ),
             self._time_and_run_task(self.build_relation_info(sender, target), "relation_info"),
-            self._time_and_run_task(self.build_memory_block(message_list_before_short, target), "memory_block"),
+            self._time_and_run_task(self.build_memory_block(chat_talking_prompt_short, target), "memory_block"),
             self._time_and_run_task(
                 self.build_tool_info(chat_talking_prompt_short, sender, target, enable_tool=enable_tool), "tool_info"
             ),
-            self._time_and_run_task(self.get_prompt_info(chat_talking_prompt_short, reply_to), "prompt_info"),
+            self._time_and_run_task(self.get_prompt_info(chat_talking_prompt_short, sender, target), "prompt_info"),
             self._time_and_run_task(
                 PromptUtils.build_cross_context(chat_id, target_user_info, global_config.personality.prompt_mode),
                 "cross_context",
@@ -1034,14 +1047,14 @@ class DefaultReplyer:
         reason: str,
         reply_to: str,
         reply_message: Optional[Dict[str, Any]] = None,
-    ) -> Tuple[str, List[int]]:  # sourcery skip: merge-else-if-into-elif, remove-redundant-if
+    ) -> str:  # sourcery skip: merge-else-if-into-elif, remove-redundant-if
         chat_stream = self.chat_stream
         chat_id = chat_stream.stream_id
         is_group_chat = bool(chat_stream.group_info)
-
+            
         if reply_message:
-            sender = reply_message.get("sender", "")
-            target = reply_message.get("target", "")
+            sender = reply_message.get("sender")
+            target = reply_message.get("target")
         else:
             sender, target = self._parse_reply_target(reply_to)
 
@@ -1206,7 +1219,6 @@ class DefaultReplyer:
         start_time = time.time()
         from src.plugins.built_in.knowledge.lpmm_get_knowledge import SearchKnowledgeFromLPMMTool
 
-
         logger.debug(f"获取知识库内容，元消息：{message[:30]}...，消息长度: {len(message)}")
         # 从LPMM知识库获取知识
         try:
@@ -1253,16 +1265,11 @@ class DefaultReplyer:
             logger.error(f"获取知识库内容时发生异常: {str(e)}")
             return ""
 
-    async def build_relation_info(self, reply_to: str = ""):
+    async def build_relation_info(self, sender: str, target: str):
         if not global_config.relationship.enable_relationship:
             return ""
 
         relationship_fetcher = relationship_fetcher_manager.get_fetcher(self.chat_stream.stream_id)
-        if not reply_to:
-            return ""
-        sender, text = self._parse_reply_target(reply_to)
-        if not sender or not text:
-            return ""
 
         # 获取用户ID
         person_info_manager = get_person_info_manager()
diff --git a/src/config/api_ada_configs.py b/src/config/api_ada_configs.py
index 13798f2a5..b74f1b558 100644
--- a/src/config/api_ada_configs.py
+++ b/src/config/api_ada_configs.py
@@ -106,8 +106,7 @@ class ModelTaskConfig(ValidatedConfigBase):
     # 必需配置项
     utils: TaskConfig = Field(..., description="组件模型配置")
     utils_small: TaskConfig = Field(..., description="组件小模型配置")
-    replyer_1: TaskConfig = Field(..., description="normal_chat首要回复模型模型配置")
-    replyer_2: TaskConfig = Field(..., description="normal_chat次要回复模型配置")
+    replyer: TaskConfig = Field(..., description="normal_chat首要回复模型模型配置")
     maizone: TaskConfig = Field(..., description="maizone专用模型")
     emotion: TaskConfig = Field(..., description="情绪模型配置")
     vlm: TaskConfig = Field(..., description="视觉语言模型配置")
diff --git a/src/config/official_configs.py b/src/config/official_configs.py
index 204fc60ac..67c2d4b6b 100644
--- a/src/config/official_configs.py
+++ b/src/config/official_configs.py
@@ -258,7 +258,6 @@ class MessageReceiveConfig(ValidatedConfigBase):
 class NormalChatConfig(ValidatedConfigBase):
     """普通聊天配置类"""
 
-    willing_mode: str = Field(default="classical", description="意愿模式")
 
 
 class ExpressionRule(ValidatedConfigBase):
diff --git a/src/main.py b/src/main.py
index 92cea1578..e9b7d48b8 100644
--- a/src/main.py
+++ b/src/main.py
@@ -189,6 +189,7 @@ MoFox_Bot(第三方修改版)
         get_emoji_manager().initialize()
         logger.info("表情包管理器初始化成功")
 
+
         # 启动情绪管理器
         await mood_manager.start()
         logger.info("情绪管理器初始化成功")
diff --git a/src/mais4u/mais4u_chat/s4u_stream_generator.py b/src/mais4u/mais4u_chat/s4u_stream_generator.py
index 607470cd2..cf102a799 100644
--- a/src/mais4u/mais4u_chat/s4u_stream_generator.py
+++ b/src/mais4u/mais4u_chat/s4u_stream_generator.py
@@ -13,12 +13,29 @@ logger = get_logger("s4u_stream_generator")
 
 class S4UStreamGenerator:
     def __init__(self):
-        # 使用LLMRequest替代AsyncOpenAIClient
-        self.llm_request = LLMRequest(
-            model_set=model_config.model_task_config.replyer, 
-            request_type="s4u_replyer"
-        )
-        
+        replyer_config = model_config.model_task_config.replyer
+        model_to_use = replyer_config.model_list[0]
+        model_info = model_config.get_model_info(model_to_use)
+        if not model_info:
+            logger.error(f"模型 {model_to_use} 在配置中未找到")
+            raise ValueError(f"模型 {model_to_use} 在配置中未找到")
+        provider_name = model_info.api_provider
+        provider_info = model_config.get_provider(provider_name)
+        if not provider_info:
+            logger.error("`replyer` 找不到对应的Provider")
+            raise ValueError("`replyer` 找不到对应的Provider")
+
+        api_key = provider_info.api_key
+        base_url = provider_info.base_url
+
+        if not api_key:
+            logger.error(f"{provider_name}没有配置API KEY")
+            raise ValueError(f"{provider_name}没有配置API KEY")
+
+        self.client_1 = AsyncOpenAIClient(api_key=api_key, base_url=base_url)
+        self.model_1_name = model_to_use
+        self.replyer_config = replyer_config
+
         self.current_model_name = "unknown model"
         self.partial_response = ""
 
@@ -83,8 +100,18 @@ class S4UStreamGenerator:
             f"{self.current_model_name}思考:{message_txt[:30] + '...' if len(message_txt) > 30 else message_txt}"
         )  # noqa: E501
 
-        # 使用LLMRequest进行流式生成
-        async for chunk in self._generate_response_with_llm_request(prompt):
+        current_client = self.client_1
+        self.current_model_name = self.model_1_name
+
+        extra_kwargs = {}
+        if self.replyer_config.get("enable_thinking") is not None:
+            extra_kwargs["enable_thinking"] = self.replyer_config.get("enable_thinking")
+        if self.replyer_config.get("thinking_budget") is not None:
+            extra_kwargs["thinking_budget"] = self.replyer_config.get("thinking_budget")
+
+        async for chunk in self._generate_response_with_model(
+            prompt, current_client, self.current_model_name, **extra_kwargs
+        ):
             yield chunk
 
     async def _generate_response_with_llm_request(self, prompt: str) -> AsyncGenerator[str, None]:
diff --git a/src/plugin_system/apis/generator_api.py b/src/plugin_system/apis/generator_api.py
index b693350b0..4ef6d1a53 100644
--- a/src/plugin_system/apis/generator_api.py
+++ b/src/plugin_system/apis/generator_api.py
@@ -73,6 +73,7 @@ async def generate_reply(
     chat_stream: Optional[ChatStream] = None,
     chat_id: Optional[str] = None,
     action_data: Optional[Dict[str, Any]] = None,
+    reply_to: str = "",
     reply_message: Optional[Dict[str, Any]] = None,
     extra_info: str = "",
     reply_reason: str = "",
@@ -92,7 +93,8 @@ async def generate_reply(
         chat_stream: 聊天流对象（优先）
         chat_id: 聊天ID（备用）
         action_data: 动作数据（向下兼容，包含reply_to和extra_info）
-        reply_message: 回复的消息对象
+        reply_to: 回复对象，格式为 "发送者:消息内容"
+        reply_message: 回复的原始消息
         extra_info: 额外信息，用于补充上下文
         reply_reason: 回复原因
         available_actions: 可用动作
@@ -133,6 +135,7 @@ async def generate_reply(
             reply_reason=reply_reason,
             from_plugin=from_plugin,
             stream_id=chat_stream.stream_id if chat_stream else chat_id,
+            reply_message=reply_message,
         )
         if not success:
             logger.warning("[GeneratorAPI] 回复生成失败")
@@ -188,7 +191,6 @@ async def rewrite_reply(
         chat_id: 聊天ID（备用）
         enable_splitter: 是否启用消息分割器
         enable_chinese_typo: 是否启用错字生成器
-        model_set_with_weight: 模型配置列表，每个元素为 (TaskConfig, weight) 元组
         raw_reply: 原始回复内容
         reason: 回复原因
         reply_to: 回复对象
diff --git a/src/plugin_system/apis/send_api.py b/src/plugin_system/apis/send_api.py
index 873e19880..91645105a 100644
--- a/src/plugin_system/apis/send_api.py
+++ b/src/plugin_system/apis/send_api.py
@@ -32,7 +32,7 @@ import traceback
 import time
 import difflib
 import asyncio
-from typing import Optional, Union, Dict
+from typing import Optional, Union, Dict, Any
 from src.common.logger import get_logger
 
 # 导入依赖
@@ -87,8 +87,9 @@ async def _send_to_target(
     stream_id: str,
     display_message: str = "",
     typing: bool = False,
+    reply_to: str = "",
     set_reply: bool = False,
-    reply_message: Optional[Dict[str, Any]] = None,
+    reply_to_message: Optional[Dict[str, Any]] = None,
     storage_message: bool = True,
     show_log: bool = True,
     selected_expressions:List[int] = None,
@@ -109,10 +110,9 @@ async def _send_to_target(
         bool: 是否发送成功
     """
     try:
-        if set_reply and not reply_message:
-            logger.warning("[SendAPI] 使用引用回复，但未提供回复消息")
-            return False
-        
+        if reply_to:
+            logger.warning("[SendAPI] 在0.10.0, reply_to 参数已弃用，请使用 reply_to_message 参数")
+
         if show_log:
             logger.debug(f"[SendAPI] 发送{message_type}消息到 {stream_id}")
 
@@ -139,16 +139,13 @@ async def _send_to_target(
         # 创建消息段
         message_segment = Seg(type=message_type, data=content)  # type: ignore
 
-        if reply_message:
-            anchor_message = message_dict_to_message_recv(reply_message)
-            if anchor_message:
-                anchor_message.update_chat_stream(target_stream)
-                assert anchor_message.message_info.user_info, "用户信息缺失"
-                reply_to_platform_id = (
-                    f"{anchor_message.message_info.platform}:{anchor_message.message_info.user_info.user_id}"
-                )            
+        if reply_to_message:
+            anchor_message = MessageRecv(message_dict=reply_to_message)
+            anchor_message.update_chat_stream(target_stream)
+            reply_to_platform_id = (
+                f"{anchor_message.message_info.platform}:{anchor_message.message_info.user_info.user_id}"
+            )            
         else:
-            reply_to_platform_id = ""
             anchor_message = None
 
         # 构建发送消息对象
@@ -249,8 +246,9 @@ async def text_to_stream(
     text: str,
     stream_id: str,
     typing: bool = False,
+    reply_to: str = "",
+    reply_to_message: Optional[Dict[str, Any]] = None,
     set_reply: bool = False,
-    reply_message: Optional[Dict[str, Any]] = None,
     storage_message: bool = True,
     selected_expressions:List[int] = None,
 ) -> bool:
@@ -272,14 +270,15 @@ async def text_to_stream(
         stream_id,
         "",
         typing,
+        reply_to,
         set_reply=set_reply,
-        reply_message=reply_message,
+        reply_to_message=reply_to_message,
         storage_message=storage_message,
         selected_expressions=selected_expressions,
     )
 
 
-async def emoji_to_stream(emoji_base64: str, stream_id: str, storage_message: bool = True, set_reply: bool = False,reply_message: Optional[Dict[str, Any]] = None) -> bool:
+async def emoji_to_stream(emoji_base64: str, stream_id: str, storage_message: bool = True, set_reply: bool = False) -> bool:
     """向指定流发送表情包
 
     Args:
@@ -290,10 +289,10 @@ async def emoji_to_stream(emoji_base64: str, stream_id: str, storage_message: bo
     Returns:
         bool: 是否发送成功
     """
-    return await _send_to_target("emoji", emoji_base64, stream_id, "", typing=False, storage_message=storage_message, set_reply=set_reply,reply_message=reply_message)
+    return await _send_to_target("emoji", emoji_base64, stream_id, "", typing=False, storage_message=storage_message, set_reply=set_reply)
 
 
-async def image_to_stream(image_base64: str, stream_id: str, storage_message: bool = True, set_reply: bool = False,reply_message: Optional[Dict[str, Any]] = None) -> bool:
+async def image_to_stream(image_base64: str, stream_id: str, storage_message: bool = True, set_reply: bool = False) -> bool:
     """向指定流发送图片
 
     Args:
@@ -304,11 +303,11 @@ async def image_to_stream(image_base64: str, stream_id: str, storage_message: bo
     Returns:
         bool: 是否发送成功
     """
-    return await _send_to_target("image", image_base64, stream_id, "", typing=False, storage_message=storage_message, set_reply=set_reply,reply_message=reply_message)
+    return await _send_to_target("image", image_base64, stream_id, "", typing=False, storage_message=storage_message, set_reply=set_reply)
 
 
 async def command_to_stream(
-    command: Union[str, dict], stream_id: str, storage_message: bool = True, display_message: str = "", set_reply: bool = False,reply_message: Optional[Dict[str, Any]] = None
+    command: Union[str, dict], stream_id: str, storage_message: bool = True, display_message: str = "", set_reply: bool = False
 ) -> bool:
     """向指定流发送命令
 
@@ -331,7 +330,8 @@ async def custom_to_stream(
     stream_id: str,
     display_message: str = "",
     typing: bool = False,
-    reply_message: Optional[Dict[str, Any]] = None,
+    reply_to: str = "",
+    reply_to_message: Optional[Dict[str, Any]] = None,
     set_reply: bool = False,
     storage_message: bool = True,
     show_log: bool = True,
@@ -356,7 +356,8 @@ async def custom_to_stream(
         stream_id=stream_id,
         display_message=display_message,
         typing=typing,
-        reply_message=reply_message,
+        reply_to=reply_to,
+        reply_to_message=reply_to_message,
         set_reply=set_reply,
         storage_message=storage_message,
         show_log=show_log,
diff --git a/src/plugins/built_in/maizone_refactored/services/content_service.py b/src/plugins/built_in/maizone_refactored/services/content_service.py
index bfa12cd32..46f6017f0 100644
--- a/src/plugins/built_in/maizone_refactored/services/content_service.py
+++ b/src/plugins/built_in/maizone_refactored/services/content_service.py
@@ -48,7 +48,7 @@ class ContentService:
         try:
             # 获取模型配置
             models = llm_api.get_available_models()
-            text_model = str(self.get_config("models.text_model", "replyer_1"))
+            text_model = str(self.get_config("models.text_model", "replyer"))
             model_config = models.get(text_model)
 
             if not model_config:
@@ -278,7 +278,7 @@ class ContentService:
         try:
             # 获取模型配置
             models = llm_api.get_available_models()
-            text_model = str(self.get_config("models.text_model", "replyer_1"))
+            text_model = str(self.get_config("models.text_model", "replyer"))
             model_config = models.get(text_model)
 
             if not model_config:
diff --git a/template/model_config_template.toml b/template/model_config_template.toml
index eadebb00c..4ac985ccc 100644
--- a/template/model_config_template.toml
+++ b/template/model_config_template.toml
@@ -1,5 +1,5 @@
 [inner]
-version = "1.2.9"
+version = "1.3.0"
 
 # 配置文件版本号迭代规则同bot_config.toml