refactor some llm api using openai api format (#1692)

### What problem does this PR solve?

refactor some llm api using openai api format

### Type of change

- [x] Refactoring

---------

Co-authored-by: Zhedong Cen <cenzhedong2@126.com>

rag/llm/cv_model.py

@@ -378,7 +378,7 @@ class OllamaCV(Base):
     def chat(self, system, history, gen_conf, image=""):
         if system:
             history[-1]["content"] = system + history[-1]["content"] + "user query: " + history[-1]["content"]
-            
+
         try:
             for his in history:
                 if his["role"] == "user":
@@ -433,27 +433,16 @@ class OllamaCV(Base):
         yield 0
 
 
-class LocalAICV(Base):
+class LocalAICV(GptV4):
     def __init__(self, key, model_name, base_url, lang="Chinese"):
+        if not base_url:
+            raise ValueError("Local cv model url cannot be None")
+        if base_url.split("/")[-1] != "v1":
+            base_url = os.path.join(base_url, "v1")
         self.client = OpenAI(api_key="empty", base_url=base_url)
         self.model_name = model_name.split("___")[0]
         self.lang = lang
 
-    def describe(self, image, max_tokens=300):
-        b64 = self.image2base64(image)
-        prompt = self.prompt(b64)
-        for i in range(len(prompt)):
-            for c in prompt[i]["content"]:
-                if "text" in c:
-                    c["type"] = "text"
-
-        res = self.client.chat.completions.create(
-            model=self.model_name,
-            messages=prompt,
-            max_tokens=max_tokens,
-        )
-        return res.choices[0].message.content.strip(), res.usage.total_tokens
-
 
 class XinferenceCV(Base):
     def __init__(self, key, model_name="", lang="Chinese", base_url=""):
@@ -549,60 +538,19 @@ class GeminiCV(Base):
         yield response._chunks[-1].usage_metadata.total_token_count
 
 
-class OpenRouterCV(Base):
+class OpenRouterCV(GptV4):
     def __init__(
         self,
         key,
         model_name,
         lang="Chinese",
-        base_url="https://openrouter.ai/api/v1/chat/completions",
+        base_url="https://openrouter.ai/api/v1",
     ):
+        if not base_url:
+            base_url = "https://openrouter.ai/api/v1"
+        self.client = OpenAI(api_key=key, base_url=base_url)
         self.model_name = model_name
         self.lang = lang
-        self.base_url = "https://openrouter.ai/api/v1/chat/completions"
-        self.key = key
-
-    def describe(self, image, max_tokens=300):
-        b64 = self.image2base64(image)
-        response = requests.post(
-            url=self.base_url,
-            headers={
-                "Authorization": f"Bearer {self.key}",
-            },
-            data=json.dumps(
-                {
-                    "model": self.model_name,
-                    "messages": self.prompt(b64),
-                    "max_tokens": max_tokens,
-                }
-            ),
-        )
-        response = response.json()
-        return (
-            response["choices"][0]["message"]["content"].strip(),
-            response["usage"]["total_tokens"],
-        )
-
-    def prompt(self, b64):
-        return [
-            {
-                "role": "user",
-                "content": [
-                    {
-                        "type": "image_url",
-                        "image_url": {"url": f"data:image/jpeg;base64,{b64}"},
-                    },
-                    {
-                        "type": "text",
-                        "text": (
-                            "请用中文详细描述一下图中的内容，比如时间，地点，人物，事情，人物心情等，如果有数据请提取出数据。"
-                            if self.lang.lower() == "chinese"
-                            else "Please describe the content of this picture, like where, when, who, what happen. If it has number data, please extract them out."
-                        ),
-                    },
-                ],
-            }
-        ]
 
 
 class LocalCV(Base):
@@ -675,12 +623,12 @@ class NvidiaCV(Base):
         ]
 
 
-class LmStudioCV(LocalAICV):
+class LmStudioCV(GptV4):
     def __init__(self, key, model_name, base_url, lang="Chinese"):
         if not base_url:
             raise ValueError("Local llm url cannot be None")
-        if base_url.split('/')[-1] != 'v1':
-            self.base_url = os.path.join(base_url,'v1')
-        self.client = OpenAI(api_key="lm-studio", base_url=self.base_url)
+        if base_url.split("/")[-1] != "v1":
+            base_url = os.path.join(base_url, "v1")
+        self.client = OpenAI(api_key="lm-studio", base_url=base_url)
         self.model_name = model_name
         self.lang = lang