Update app.py

app.py

@@ -101,15 +101,12 @@ def refresh_models():
     """
     global text_models, free_text_models
     global embedding_models, free_embedding_models
-
-    text_models = get_all_models(
-        FREE_MODEL_TEST_KEY,
-        "chat"
-    )
-    embedding_models = get_all_models(
-        FREE_MODEL_TEST_KEY,
-        "embedding"
-    ) + get_all_models(FREE_MODEL_TEST_KEY, "reranker")
+    
+    text_models = get_all_models(FREE_MODEL_TEST_KEY, "chat")
+    embedding_models = {
+        sub_type: get_all_models(FREE_MODEL_TEST_KEY, sub_type)
+        for sub_type in ["embedding", "reranker"]
+    }
     free_text_models = []
     free_embedding_models = []
 
@@ -123,9 +120,7 @@ def refresh_models():
                 model
             ): model for model in text_models
         }
-        for future in concurrent.futures.as_completed(
-            future_to_model
-        ):
+        for future in concurrent.futures.as_completed(future_to_model):
             model = future_to_model[future]
             try:
                 is_free = future.result()
@@ -140,13 +135,10 @@ def refresh_models():
         future_to_model = {
             executor.submit(
                 test_embedding_model_availability,
-                FREE_MODEL_TEST_KEY,
-                model
+                FREE_MODEL_TEST_KEY, model
             ): model for model in embedding_models
         }
-        for future in concurrent.futures.as_completed(
-            future_to_model
-        ):
+        for future in concurrent.futures.as_completed(future_to_model):
             model = future_to_model[future]
             try:
                 is_free = future.result()
@@ -164,9 +156,6 @@ def test_embedding_model_availability(api_key, model_name):
     """
     测试指定的向量模型是否可用。
     """
-    logging.info(
-        f"测试向量模型 {model_name}"
-    )
     headers = {
         "Authorization": f"Bearer {api_key}",
         "Content-Type": "application/json"
@@ -181,34 +170,10 @@ def test_embedding_model_availability(api_key, model_name):
             },
             timeout=10
         )
-        
-        if response.status_code == 200:
-            return True
-        elif response.status_code == 429:
-            logging.warning(
-                f"测试向量模型 {model_name} 可用性时遇到频率限制，"
-                f"API Key：{api_key}，响应：{response.text}"
-            )
+        if response.status_code == 429 or response.status_code == 200:
             return True
-        elif response.status_code == 400:
-            try:
-                error_message = response.json().get("error", {}).get("message", "")
-                if (
-                    "maximum context length" in error_message or
-                    "This model's maximum context length" in error_message
-                ):
-                    logging.warning(
-                        f"测试向量模型 {model_name} 可用性失败，"
-                        f"API Key：{api_key}，错误信息：模型不支持大于1的输入"
-                    )
-                    return False
-                else:
-                    return False
-            except (KeyError, TypeError, json.JSONDecodeError):
-                return False
         else:
             return False
-
     except requests.exceptions.RequestException as e:
         logging.error(
             f"测试向量模型 {model_name} 可用性失败，"