Feat: Add qwq model support to Tongyi-Qianwen factory (#5981)

### What problem does this PR solve?

add qwq model support to Tongyi-Qianwen factory
https://github.com/infiniflow/ragflow/issues/5869

### Type of change

- [x] New Feature (non-breaking change which adds functionality)


![image](https://github.com/user-attachments/assets/49f5c6a0-ecaf-41dd-a23a-2009f854d62c)


![image](https://github.com/user-attachments/assets/93ffa303-920e-4942-8188-bcd6b7209204)


![1741774779](https://github.com/user-attachments/assets/25f2fd1d-8640-4df0-9a08-78ee9daaa8fe)


![image](https://github.com/user-attachments/assets/4763cf6c-1f76-43c4-80ee-74dfd666a184)

Co-authored-by: zhaozhicheng <zhicheng.zhao@fastonetech.com>

conf/llm_factories.json

@@ -134,6 +134,18 @@
                     "max_tokens": 32768,
                     "model_type": "chat"
                 },
+                {
+                    "llm_name": "qwq-32b",
+                    "tags": "LLM,CHAT,128k",
+                    "max_tokens": 131072,
+                    "model_type": "chat"
+                },
+                {
+                    "llm_name": "qwq-plus",
+                    "tags": "LLM,CHAT,128k",
+                    "max_tokens": 131072,
+                    "model_type": "chat"
+                },
                 {
                     "llm_name": "qwen-long",
                     "tags": "LLM,CHAT,10000K",
@@ -3259,7 +3271,7 @@
                     "tags": "TEXT EMBEDDING,32000",
                     "max_tokens": 32000,
                     "model_type": "embedding"
-                },                
+                },
                 {
                     "llm_name": "rerank-1",
                     "tags": "RE-RANK, 8000",

rag/llm/chat_model.py

@@ -268,13 +268,13 @@ class QWenChat(Base):
         import dashscope
         dashscope.api_key = key
         self.model_name = model_name
-        if model_name.lower().find("deepseek") >= 0:
+        if self.is_reasoning_model(self.model_name):
             super().__init__(key, model_name, "https://dashscope.aliyuncs.com/compatible-mode/v1")
 
     def chat(self, system, history, gen_conf):
         if "max_tokens" in gen_conf:
             del gen_conf["max_tokens"]
-        if self.model_name.lower().find("deepseek") >= 0:
+        if self.is_reasoning_model(self.model_name):
             return super().chat(system, history, gen_conf)
 
         stream_flag = str(os.environ.get('QWEN_CHAT_BY_STREAM', 'true')).lower() == 'true'
@@ -348,11 +348,19 @@ class QWenChat(Base):
     def chat_streamly(self, system, history, gen_conf):
         if "max_tokens" in gen_conf:
             del gen_conf["max_tokens"]
-        if self.model_name.lower().find("deepseek") >= 0:
+        if self.is_reasoning_model(self.model_name):
             return super().chat_streamly(system, history, gen_conf)
 
         return self._chat_streamly(system, history, gen_conf)
 
+    @staticmethod
+    def is_reasoning_model(model_name: str) -> bool:
+        return any([
+            model_name.lower().find("deepseek") >= 0,
+            model_name.lower().find("qwq") >= 0 and model_name.lower() != 'qwq-32b-preview',
+        ])
+
+
 
 class ZhipuChat(Base):
     def __init__(self, key, model_name="glm-3-turbo", **kwargs):
@@ -740,7 +748,7 @@ class BedrockChat(Base):
         self.bedrock_sk = json.loads(key).get('bedrock_sk', '')
         self.bedrock_region = json.loads(key).get('bedrock_region', '')
         self.model_name = model_name
-        
+
         if self.bedrock_ak == '' or self.bedrock_sk == '' or self.bedrock_region == '':
             # Try to create a client using the default credentials (AWS_PROFILE, AWS_DEFAULT_REGION, etc.)
             self.client = boto3.client('bedrock-runtime')