diff --git a/backend/package/yuxi/agents/buildin/deep_agent/context.py b/backend/package/yuxi/agents/buildin/deep_agent/context.py
index a29372df..8a2254cf 100644
--- a/backend/package/yuxi/agents/buildin/deep_agent/context.py
+++ b/backend/package/yuxi/agents/buildin/deep_agent/context.py
@@ -22,7 +22,7 @@ class DeepContext(BaseContext):
     )
 
     subagents_model: Annotated[str, {"__template_metadata__": {"kind": "llm"}}] = field(
-        default="siliconflow/Pro/deepseek-ai/DeepSeek-V3.2",
+        default="siliconflow-cn:deepseek-ai/DeepSeek-V4-Flash",
         metadata={
             "name": "Sub-agent Model",
             "description": "子智能体的默认模型，会被子智能体的配置覆盖。",
diff --git a/backend/package/yuxi/agents/models.py b/backend/package/yuxi/agents/models.py
index fb77a1a7..aef96c1a 100644
--- a/backend/package/yuxi/agents/models.py
+++ b/backend/package/yuxi/agents/models.py
@@ -13,7 +13,7 @@ from yuxi.utils.logging_config import logger
 def load_chat_model_v2(spec: str, **kwargs) -> BaseChatModel:
     """根据 v2 spec（provider_id:model_id）加载 LangChain 聊天模型。
 
-    v2 spec 格式使用冒号分隔，如: siliconflow-cn:Pro/deepseek-ai/DeepSeek-V3.2
+    v2 spec 格式使用冒号分隔，如: siliconflow-cn:deepseek-ai/DeepSeek-V4-Flash
     数据来源为数据库中的 model_providers 表，通过全局缓存访问。
     """
     from yuxi.services.model_cache import model_cache
diff --git a/backend/package/yuxi/config/builtin_providers.py b/backend/package/yuxi/config/builtin_providers.py
index 6aaef2f8..b62b60b7 100644
--- a/backend/package/yuxi/config/builtin_providers.py
+++ b/backend/package/yuxi/config/builtin_providers.py
@@ -155,7 +155,7 @@ BUILTIN_PROVIDERS: list[dict[str, Any]] = [
         "embedding_models_endpoint": "https://api.siliconflow.cn/v1/models?sub_type=embedding",
         "rerank_models_endpoint": "https://api.siliconflow.cn/v1/models?sub_type=reranker",
         "enabled_models": [
-            {"id": "Pro/deepseek-ai/DeepSeek-V3.2", "type": "chat", "display_name": "Pro/deepseek-ai/DeepSeek-V3.2"},
+            {"id": "deepseek-ai/DeepSeek-V4-Flash", "type": "chat", "display_name": "deepseek-ai/DeepSeek-V4-Flash"},
             {"id": "Pro/MiniMaxAI/MiniMax-M2.5", "type": "chat", "display_name": "Pro/MiniMaxAI/MiniMax-M2.5"},
             {
                 "id": "Pro/BAAI/bge-m3",
diff --git a/backend/package/yuxi/models/chat.py b/backend/package/yuxi/models/chat.py
index 22d19f62..2f5381cf 100644
--- a/backend/package/yuxi/models/chat.py
+++ b/backend/package/yuxi/models/chat.py
@@ -104,7 +104,7 @@ class GeneralResponse:
 def select_model_v2(spec: str) -> OpenAIBase:
     """根据 v2 spec（provider_id:model_id）选择聊天模型。
 
-    v2 spec 格式使用冒号分隔，如: siliconflow-cn:Pro/deepseek-ai/DeepSeek-V3.2
+    v2 spec 格式使用冒号分隔，如: siliconflow-cn:deepseek-ai/DeepSeek-V4-Flash
     数据来源为数据库中的 model_providers 表，通过全局缓存访问。
     """
     from yuxi.services.model_cache import model_cache
diff --git a/docs/advanced/configuration.md b/docs/advanced/configuration.md
index d1f2d455..58186b4d 100644
--- a/docs/advanced/configuration.md
+++ b/docs/advanced/configuration.md
@@ -45,13 +45,13 @@ class Config(BaseModel):
     enable_content_guard: bool = Field(default=False, description="是否启用内容审查")
 
     # 模型配置
-    default_model: str = Field(default="siliconflow/Pro/deepseek-ai/DeepSeek-V3.2")
-    embed_model: str = Field(default="siliconflow/BAAI/bge-m3")
+    default_model: str = Field(default="siliconflow-cn:deepseek-ai/DeepSeek-V4-Flash")
+    embed_model: str = Field(default="siliconflow-cn:BAAI/bge-m3")
 ```
 
 ### 模型配置
 
-模型配置独立管理，支持多种模型提供商：
+模型配置独立管理，支持多种模型提供商:
 
 ```python
 DEFAULT_CHAT_MODEL_PROVIDERS: dict[str, ChatModelProvider] = {
@@ -59,9 +59,9 @@ DEFAULT_CHAT_MODEL_PROVIDERS: dict[str, ChatModelProvider] = {
         name="SiliconFlow",
         url="https://cloud.siliconflow.cn/models",
         base_url="https://api.siliconflow.cn/v1",
-        default="deepseek-ai/DeepSeek-V3.2",
+        default="deepseek-ai/DeepSeek-V4-Flash",
         env="SILICONFLOW_API_KEY",
-        models=["deepseek-ai/DeepSeek-V3.2", "Qwen/Qwen3-235B-A22B-Instruct-2507"],
+        models=["deepseek-ai/DeepSeek-V4-Flash", "Qwen/Qwen3-235B-A22B-Instruct-2507"],
     ),
     # 其他提供商...
 }