Feat: add LongCat-Flash-Chat (#9973)

### What problem does this PR solve?

Add LongCat-Flash-Chat from Meituan, deepseek v3.1 from SiliconFlow,
kimi-k2-09-05-preview and kimi-k2-turbo-preview from Moonshot.

### Type of change

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

Author: yongtenglei

conf/llm_factories.json

@@ -755,6 +755,20 @@
                     "model_type": "chat",
                     "is_tools": true
                 },
+                {
+                    "llm_name": "kimi-k2-0905-preview",
+                    "tags": "LLM,CHAT,256k",
+                    "max_tokens": 262144,
+                    "model_type": "chat",
+                    "is_tools": true
+                },
+                {
+                    "llm_name": "kimi-k2-turbo-preview",
+                    "tags": "LLM,CHAT,256k",
+                    "max_tokens": 262144,
+                    "model_type": "chat",
+                    "is_tools": true
+                },
                 {
                     "llm_name": "kimi-latest",
                     "tags": "LLM,CHAT,8k,32k,128k",
@@ -2794,6 +2808,20 @@
                     "model_type": "chat",
                     "is_tools": true
                 },
+                {
+                    "llm_name": "Pro/deepseek-ai/DeepSeek-V3.1",
+                    "tags": "LLM,CHAT,160k",
+                    "max_tokens": 160000,
+                    "model_type": "chat",
+                    "is_tools": true
+                },
+                {
+                    "llm_name": "deepseek-ai/DeepSeek-V3.1",
+                    "tags": "LLM,CHAT,160",
+                    "max_tokens": 160000,
+                    "model_type": "chat",
+                    "is_tools": true
+                },
                 {
                     "llm_name": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
                     "tags": "LLM,CHAT,32k",
@@ -4448,6 +4476,21 @@
                     "is_tools": false
                 }
             ]
+        },
+        {
+            "name": "Meituan",
+            "logo": "",
+            "tags": "LLM",
+            "status": "1",
+            "llm": [
+                {
+                    "llm_name": "LongCat-Flash-Chat",
+                    "tags": "LLM,CHAT,8000",
+                    "max_tokens": 8000,
+                    "model_type": "chat",
+                    "is_tools": true
+                }
+            ]
         }
     ]
 }

rag/llm/chat_model.py

@@ -155,7 +155,10 @@ def _chat(self, history, gen_conf, **kwargs):
     def _chat_streamly(self, history, gen_conf, **kwargs):
         logging.info("[HISTORY STREAMLY]" + json.dumps(history, ensure_ascii=False, indent=4))
         reasoning_start = False
-        response = self.client.chat.completions.create(model=self.model_name, messages=history, stream=True, **gen_conf, stop=kwargs.get("stop"))
+        if kwargs.get("stop") or "stop" in gen_conf:
+            response = self.client.chat.completions.create(model=self.model_name, messages=history, stream=True, **gen_conf, stop=kwargs.get("stop"))
+        else:
+            response = self.client.chat.completions.create(model=self.model_name, messages=history, stream=True, **gen_conf)
         for resp in response:
             if not resp.choices:
                 continue
@@ -1353,6 +1356,15 @@ def __init__(self, key, model_name, base_url="https://api.302.ai/v1", **kwargs):
         super().__init__(key, model_name, base_url, **kwargs)
 
 
+class MeituanChat(Base):
+    _FACTORY_NAME = "Meituan"
+
+    def __init__(self, key, model_name, base_url="https://api.longcat.chat/openai", **kwargs):
+        if not base_url:
+            base_url = "https://api.longcat.chat/openai"
+        super().__init__(key, model_name, base_url, **kwargs)
+
+
 class LiteLLMBase(ABC):
     _FACTORY_NAME = ["Tongyi-Qianwen", "Bedrock", "Moonshot", "xAI", "DeepInfra", "Groq", "Cohere", "Gemini", "DeepSeek", "NVIDIA", "TogetherAI", "Anthropic", "Ollama"]
 

web/src/assets/svg/llm/longcat.svg

@@ -54,6 +54,7 @@ export enum LLMFactory {
   DeepInfra = 'DeepInfra',
   Grok = 'Grok',
   XAI = 'xAI',
+  Meituan = 'Meituan',
 }
 
 // Please lowercase the file name
@@ -113,4 +114,5 @@ export const IconMap = {
   [LLMFactory.DeepInfra]: 'deepinfra',
   [LLMFactory.Grok]: 'grok',
   [LLMFactory.XAI]: 'xai',
+  [LLMFactory.Meituan]: 'longcat',
 };