Update app.py

app.py

@@ -67,7 +67,7 @@ def test_model_availability(api_key, model_name):
                                  json={
                                      "model": model_name,
                                      "messages": [{"role": "user", "content": "hi"}],
-                                     "max_tokens": 10,
+                                     "max_tokens": 5,
                                      "stream": False
                                  },
                                  timeout=10)
@@ -359,7 +359,6 @@ def handsome_chat_completions():
                 first_token_time = first_chunk_time - start_time if first_chunk_time else 0
                 total_time = end_time - start_time
 
-                # 处理流式响应，逐行解析 JSON
                 prompt_tokens = 0
                 completion_tokens = 0
                 response_content = ""
@@ -371,7 +370,6 @@ def handsome_chat_completions():
                         try:
                             response_json = json.loads(line)
                             
-                            # 提取信息，这里只累加 completion_tokens 和 content
                             if "usage" in response_json and "completion_tokens" in response_json["usage"]:
                                 completion_tokens = response_json["usage"]["completion_tokens"]
                             
@@ -384,7 +382,6 @@ def handsome_chat_completions():
                         except (KeyError, ValueError, IndexError) as e:
                             logging.error(f"解析流式响应单行 JSON 失败: {e}, 行内容: {line}")
                             
-                # 提取用户输入的内容
                 user_content = ""
                 messages = data.get("messages", [])
                 for message in messages:
@@ -392,19 +389,16 @@ def handsome_chat_completions():
                         user_content += message["content"] + " "
                 user_content = user_content.strip()
 
-                # 记录日志
                 logging.info(
                     f"使用的key: {api_key}, 提示token: {prompt_tokens}, 输出token: {completion_tokens}, 首字用时: {first_token_time:.4f}秒, 总共用时: {total_time:.4f}秒, 使用的模型: {model_name}, 用户的内容: {user_content}, 输出的内容: {response_content}"
                 )
             return Response(stream_with_context(generate()), content_type=response.headers['Content-Type'])
         else:
-            # 非流式响应处理... (保持原样)
             response.raise_for_status()
             end_time = time.time()
             response_json = response.json()
             total_time = end_time - start_time
 
-            # 从响应中提取信息
             try:
                 prompt_tokens = response_json["usage"]["prompt_tokens"]
                 completion_tokens = response_json["usage"]["completion_tokens"]
@@ -415,7 +409,6 @@ def handsome_chat_completions():
                 completion_tokens = 0
                 response_content = ""
 
-            # 提取用户输入的内容
             user_content = ""
             messages = data.get("messages", [])
             for message in messages:
@@ -423,7 +416,6 @@ def handsome_chat_completions():
                     user_content += message["content"] + " "
             user_content = user_content.strip()
 
-            # 记录日志
             logging.info(
                 f"使用的key: {api_key}, 提示token: {prompt_tokens}, 输出token: {completion_tokens}, 首字用时: 0, 总共用时: {total_time:.4f}秒, 使用的模型: {model_name}, 用户的内容: {user_content}, 输出的内容: {response_content}"
             )
@@ -432,8 +424,6 @@ def handsome_chat_completions():
     except requests.exceptions.RequestException as e:
         return jsonify({"error": str(e)}), 500
 
-
-
 @app.route('/handsome/v1/models', methods=['GET'])
 def list_models():
     if not check_authorization(request):
@@ -533,4 +523,4 @@ if __name__ == '__main__':
     refresh_models()
     logging.info("首次刷新模型列表已手动触发执行")
 
-    app.run(debug=False, host='0.0.0.0', port=int(os.environ.get('PORT', 7860)))
+    app.run(debug=False, host='0.0.0.0', port=int(os.environ.get('PORT', 7860)))