diff --git a/.gitignore b/.gitignore
index b321fa2..c28469d 100644
--- a/.gitignore
+++ b/.gitignore
@@ -8,4 +8,5 @@ build/
 **/__pycache__
 info.sh
 tests/test.py
-claude.sh
\ No newline at end of file
+claude.sh
+*.txt
\ No newline at end of file
diff --git a/terminal_agent_x/tax.py b/terminal_agent_x/tax.py
index 64ad233..03f132c 100644
--- a/terminal_agent_x/tax.py
+++ b/terminal_agent_x/tax.py
@@ -63,10 +63,11 @@ def fetch_code(openai_key: str, model: str, prompt: str, url_option: str, chat_f
         command = f'{command} --ipv4' if model == 'claude' else command
     else:  # Linux
         command = f"curl -s --location '{url}' --header '{headers[0]}' --header '{headers[1]}' --data '{data}'"
-    # print(command)
+    print(command)
 
     try:
         res, err = run_command_with_timeout(command, 60)
+        print(res)
         # res = os.popen(command).read().encode('utf-8').decode('utf-8', 'ignore')
         if model.lower() == 'dalle':
             return json.loads(res)['data'][0]['url']
@@ -166,8 +167,7 @@ def chat(openai_key: str, model: str, url_option: str):
         if user_input == "exit":
             break
         conversation.append({"role": "user", "content": user_input})
-        response = fetch_code(openai_key, model, json.dumps(
-            conversation), url_option, True)
+        response = fetch_code(openai_key, model, json.dumps(conversation), url_option, True)
         print(f'Tax: {response}')
         conversation.append({"role": "assistant", "content": response.encode(
             'unicode-escape').decode('utf8').replace("'", "")})
@@ -175,23 +175,31 @@ def chat(openai_key: str, model: str, url_option: str):
         # print(conversation)
 
 
-def parallel_ask(data_prompts, chat_model, max_workers, output_file, **args):
-    with concurrent.futures.ThreadPoolExecutor(max_workers=max_workers) as executor:
-        future_to_prompt = {executor.submit(chat_model, prompt=prompt, **args): prompt for prompt in data_prompts}
+def parallel_ask(data_prompts, chat_mode, max_workers, output_file, model, **args):
+    with concurrent.futures.ThreadPoolExecutor(max_workers=int(max_workers)) as executor:
+        future_to_prompt = []
+        for prompt in data_prompts:
+            future_to_prompt.append(executor.submit(
+                chat_mode, prompt=prompt, model=model, **args))
+        results = []
         for future in concurrent.futures.as_completed(future_to_prompt):
-            prompt = future_to_prompt[future]
             try:
                 data = future.result()
             except Exception as exc:
-                print(f'{prompt} generated an exception: {exc}')
-            else:
-                # print(f'{prompt} generated {data}')
-                if output_file:
-                    with open(output_file, 'a', encoding='utf-8') as f:
-                        f.write(f'{prompt} : {data}\n\n')
-                    f.close()
+                data = str(type(exc))
+            results.append(data)
+    if output_file:
+        with open(output_file, 'w', encoding='utf-8') as f:
+            f.write('\n'.join(results))
+        f.close()
 
 
+def load_prompts_file(model, path: str) -> str:
+    with open(path, 'r', encoding='utf-8') as f:
+        text = f.readlines()
+    text = [line.strip() for line in text]
+    wrappers = [chat_data_wrapper(model, prompt, False) for prompt in text]
+    return wrappers
 
 
 def main() -> None:
@@ -212,6 +220,9 @@ def main() -> None:
                         help="URL for API request. Choose from ['openai_gfw', 'openai', 'claude'] or your custom url.")
     parser.add_argument('-a', '--show_all', action='store_true',
                         help='Show all contents in the response.')
+    parser.add_argument('-p', '--parallel', action='store_true',
+                        help='Parallel mode. If specified, the input file will be read line by line and the responses will be saved to the output file.')
+    parser.add_argument('--option', metavar='KEY=VALUE', action='append', help='Custom option')
     args = parser.parse_args()
 
     prompt = ' '.join(args.prompt)
@@ -232,6 +243,17 @@ def main() -> None:
     # res = get_model_response(openai_key, args.model, prompt)
     res = fetch_code(key, args.model, prompt, args.url, False)
 
+    if args.option and args.parallel:
+        custom_options = {}
+        for option in args.option:
+            key, value = option.split('=')
+            custom_options[key] = value
+
+        print(parallel_ask(data_prompts=load_prompts_file(args.model, args.input), output_file=args.output, model=args.model, **custom_options))
+        return 
+
+    # tax -i input.txt -o output.txt -m gpt-3.5-turbo -u openai_gfw -k xxx --option chat_mode=fetch_code 
+
     if args.output:
         with open(args.output, 'w', encoding='utf-8') as f:
             f.write(res)
@@ -252,5 +274,5 @@ def main() -> None:
 
 
 if __name__ == '__main__':
-    # main()
-    parallel_ask(data_prompts=['hi'], chat_model=fetch_code, max_workers=3, output_file='output.txt', openai_key='', model='gpt-3.5-turbo', url_option='openai', chat_flag=False)
+    main()
+    # parallel_ask(data_prompts=['hi'], chat_mode=fetch_code, max_workers=3, output_file='output.txt', openai_key='', model='gpt-3.5-turbo', url_option='openai', chat_flag=False)