fix: thinking display is disabled by default

2026-06-02 00:57:41 +08:00 · 2026-04-17 15:31:59 +08:00
parent 35282db9e0
commit 13370d2056
9 changed files with 25 additions and 19 deletions
--- a/README.md
+++ b/README.md
@@ -206,7 +206,7 @@ cow install-browser
  "agent_max_context_tokens": 50000,                          # Agent 模式下最大上下文 tokens，超出将自动智能压缩处理
  "agent_max_context_turns": 20,                              # Agent 模式下最大上下文记忆轮次，一问一答为一轮，超出后智能压缩处理
  "agent_max_steps": 20,                                      # Agent 模式下单次任务的最大决策步数，超出后将停止继续调用工具
-  "enable_thinking": true                                     # 是否启用深度思考，开启后 Web 端展示模型推理过程，关闭后可加速响应
+  "enable_thinking": false                                    # 是否启用深度思考，开启后 Web 端展示模型推理过程，关闭后可加速响应
 }
 ```

--- a/agent/protocol/agent_stream.py
+++ b/agent/protocol/agent_stream.py
@@ -81,20 +81,26 @@ class AgentStreamExecutor:
    def _is_thinking_enabled(self) -> bool:
        from config import conf
        channel_type = getattr(self.model, 'channel_type', '') or ''
-        return conf().get("enable_thinking", True) and channel_type == 'web'
+        return conf().get("enable_thinking", False) and channel_type == 'web'

    def _filter_think_tags(self, text: str) -> str:
        """
-        Remove <think> and </think> tags but keep the content inside.
-        Some LLM providers (e.g., MiniMax) may return thinking process wrapped in <think> tags.
-        We only remove the tags themselves, keeping the actual thinking content.
+        Handle <think>...</think> blocks in content returned by some LLM providers
+        (e.g., MiniMax).
+
+        - When thinking is enabled: remove the tags but keep the content inside.
+        - When thinking is disabled: remove both the tags and the content entirely.
        """
        if not text:
            return text
        import re
-        # Remove only the <think> and </think> tags, keep the content
-        text = re.sub(r'<think>', '', text)
-        text = re.sub(r'</think>', '', text)
+        if self._is_thinking_enabled():
+            text = re.sub(r'<think>', '', text)
+            text = re.sub(r'</think>', '', text)
+        else:
+            text = re.sub(r'<think>[\s\S]*?</think>', '', text)
+            # Also strip unclosed <think> tag at the end (streaming partial)
+            text = re.sub(r'<think>[\s\S]*$', '', text)
        return text

    def _hash_args(self, args: dict) -> str:
--- a/bridge/agent_bridge.py
+++ b/bridge/agent_bridge.py
@@ -169,7 +169,7 @@ class AgentLLMModel(LLMModel):

                # Determine thinking: respect global config, then channel_type
                from config import conf
-                global_thinking = conf().get("enable_thinking", True)
+                global_thinking = conf().get("enable_thinking", False)
                if not global_thinking:
                    kwargs['thinking'] = {"type": "disabled"}
                else:
@@ -222,7 +222,7 @@ class AgentLLMModel(LLMModel):

                # Determine thinking: respect global config, then channel_type
                from config import conf
-                global_thinking = conf().get("enable_thinking", True)
+                global_thinking = conf().get("enable_thinking", False)
                if not global_thinking:
                    kwargs['thinking'] = {"type": "disabled"}
                else:
--- a/channel/web/chat.html
+++ b/channel/web/chat.html
@@ -552,7 +552,7 @@
                                                <span class="cfg-tip" data-tip-key="config_enable_thinking_hint"><i class="fas fa-circle-question"></i></span>
                                            </label>
                                            <label class="relative inline-flex items-center cursor-pointer">
-                                                <input id="cfg-enable-thinking" type="checkbox" class="sr-only peer" checked>
+                                                <input id="cfg-enable-thinking" type="checkbox" class="sr-only peer">
                                                <div class="w-9 h-5 bg-slate-200 dark:bg-slate-700 peer-checked:bg-primary-400 rounded-full
                                                            after:content-[''] after:absolute after:top-[2px] after:left-[2px] after:bg-white
                                                            after:rounded-full after:h-4 after:w-4 after:transition-all peer-checked:after:translate-x-full"></div>
--- a/channel/web/static/js/console.js
+++ b/channel/web/static/js/console.js
@@ -2116,7 +2116,7 @@ function initConfigView(data) {
    document.getElementById('cfg-max-tokens').value = data.agent_max_context_tokens || 50000;
    document.getElementById('cfg-max-turns').value = data.agent_max_context_turns || 20;
    document.getElementById('cfg-max-steps').value = data.agent_max_steps || 20;
-    document.getElementById('cfg-enable-thinking').checked = data.enable_thinking !== false;
+    document.getElementById('cfg-enable-thinking').checked = data.enable_thinking === true;

    const pwdInput = document.getElementById('cfg-password');
    const maskedPwd = data.web_password_masked || '';
--- a/channel/web/web_channel.py
+++ b/channel/web/web_channel.py
@@ -712,7 +712,7 @@ class ChatHandler:
 class ConfigHandler:

    _RECOMMENDED_MODELS = [
-        const.MINIMAX_M2_7, const.MINIMAX_M2_5, const.MINIMAX_M2_1, const.MINIMAX_M2_1_LIGHTNING,
+        const.MINIMAX_M2_7_HIGHSPEED, const.MINIMAX_M2_7, const.MINIMAX_M2_5, const.MINIMAX_M2_1, const.MINIMAX_M2_1_LIGHTNING,
        const.GLM_5_TURBO, const.GLM_5, const.GLM_4_7,
        const.QWEN36_PLUS, const.QWEN35_PLUS, const.QWEN3_MAX,
        const.KIMI_K2_5, const.KIMI_K2,
@@ -871,7 +871,7 @@ class ConfigHandler:
                "agent_max_context_tokens": local_config.get("agent_max_context_tokens", 50000),
                "agent_max_context_turns": local_config.get("agent_max_context_turns", 20),
                "agent_max_steps": local_config.get("agent_max_steps", 20),
-                "enable_thinking": bool(local_config.get("enable_thinking", True)),
+                "enable_thinking": bool(local_config.get("enable_thinking", False)),
                "api_bases": api_bases,
                "api_keys": api_keys_masked,
                "providers": providers,
--- a/config.py
+++ b/config.py
@@ -204,7 +204,7 @@ available_setting = {
    "agent_max_context_tokens": 50000,  # Agent模式下最大上下文tokens
    "agent_max_context_turns": 20,  # Agent模式下最大上下文记忆轮次
    "agent_max_steps": 20,  # Agent模式下单次运行最大决策步数
-    "enable_thinking": True,  # Whether to enable deep thinking for web channel
+    "enable_thinking": False,  # Whether to enable deep thinking for web channel
    "knowledge": True,  # 是否开启知识库功能
 }

--- a/docs/channels/web.mdx
+++ b/docs/channels/web.mdx
@@ -12,7 +12,7 @@ Web 控制台是 CowAgent 的默认通道，启动后会自动运行，通过浏
  "channel_type": "web",
  "web_port": 9899,
  "web_password": "",
-  "enable_thinking": true
+  "enable_thinking": false
 }
 ```

@@ -22,7 +22,7 @@ Web 控制台是 CowAgent 的默认通道，启动后会自动运行，通过浏
 | `web_port` | Web 服务监听端口 | `9899` |
 | `web_password` | 访问密码，留空表示不启用密码保护 | `""` |
 | `web_session_expire_days` | 登录会话有效天数 | `30` |
-| `enable_thinking` | 是否启用深度思考，开启后 Web 端展示推理过程，关闭可加速响应 | `true` |
+| `enable_thinking` | 是否启用深度思考，开启后 Web 端展示推理过程，关闭可加速响应 | `false` |

 配置密码后，访问控制台时需先输入密码完成登录。登录状态默认保持 30 天，期间重启服务也无需重新登录。密码也支持在控制台的「配置」页面中在线修改。

--- a/docs/intro/architecture.mdx
+++ b/docs/intro/architecture.mdx
@@ -70,7 +70,7 @@ Agent 的工作空间默认位于 `~/cow` 目录，用于存储系统提示词
  "agent_max_context_tokens": 40000,
  "agent_max_context_turns": 30,
  "agent_max_steps": 15,
-  "enable_thinking": true
+  "enable_thinking": false
 }
 ```

@@ -81,5 +81,5 @@ Agent 的工作空间默认位于 `~/cow` 目录，用于存储系统提示词
 | `agent_max_context_tokens` | 最大上下文 token 数 | `50000` |
 | `agent_max_context_turns` | 最大上下文记忆轮次 | `20` |
 | `agent_max_steps` | 单次任务最大决策步数 | `20` |
-| `enable_thinking` | 是否启用深度思考，开启后 Web 端展示推理过程，关闭可加速响应 | `true` |
+| `enable_thinking` | 是否启用深度思考，开启后 Web 端展示推理过程，关闭可加速响应 | `false` |
 | `knowledge` | 是否启用个人知识库 | `true` |