update

README.md

@@ -486,9 +486,11 @@ python3 examples/openai_client.py --base_url http://127.0.0.1:8000/v1 --tool_cal
 Request GLM4V with image inputs:
 ```sh
 # request with local image file
-python3 examples/openai_client.py --base_url http://127.0.0.1:8000/v1 --tool_call --prompt 描述这张图片 --image examples/03-Confusing-Pictures.jpg
+python3 examples/openai_client.py --base_url http://127.0.0.1:8000/v1 --prompt "描述这张图片" \
+    --image examples/03-Confusing-Pictures.jpg --temp 0
 # request with image url
-python3 examples/openai_client.py --base_url http://127.0.0.1:8000/v1 --tool_call --prompt 描述这张图片 --image https://www.barnorama.com/wp-content/uploads/2016/12/03-Confusing-Pictures.jpg
+python3 examples/openai_client.py --base_url http://127.0.0.1:8000/v1 --prompt "描述这张图片" \
+    --image https://www.barnorama.com/wp-content/uploads/2016/12/03-Confusing-Pictures.jpg --temp 0
 ```
 
 With this API server as backend, ChatGLM.cpp models can be seamlessly integrated into any frontend that uses OpenAI-style API, including [mckaywrigley/chatbot-ui](https://github.com/mckaywrigley/chatbot-ui), [fuergaosi233/wechat-chatgpt](https://github.com/fuergaosi233/wechat-chatgpt), [Yidadaa/ChatGPT-Next-Web](https://github.com/Yidadaa/ChatGPT-Next-Web), and more.

examples/openai_client.py

@@ -5,12 +5,14 @@
 from openai import OpenAI
 
 parser = argparse.ArgumentParser()
-parser.add_argument("--api_key", default="Bearer chatglm-cpp-example", type=str)
-parser.add_argument("--base_url", default=None, type=str)
-parser.add_argument("--stream", action="store_true")
-parser.add_argument("--prompt", default="你好", type=str)
-parser.add_argument("--tool_call", action="store_true")
-parser.add_argument("--image", default=None, type=str)
+parser.add_argument("--api_key", default="Bearer chatglm-cpp-example", type=str, help="API key of OpenAI api server")
+parser.add_argument("--base_url", default=None, type=str, help="base url of OpenAI api server")
+parser.add_argument("--stream", action="store_true", help="enable stream generation")
+parser.add_argument("-p", "--prompt", default="你好", type=str, help="prompt to start generation with")
+parser.add_argument("--tool_call", action="store_true", help="enable function call")
+parser.add_argument("--image", default=None, type=str, help="path to the input image for visual language models")
+parser.add_argument("--temp", default=0.95, type=float, help="temperature")
+parser.add_argument("--top_p", default=0.7, type=float, help="top-p sampling")
 args = parser.parse_args()
 
 client = OpenAI(api_key=args.api_key, base_url=args.base_url)
@@ -48,13 +50,14 @@
     user_content = args.prompt
 
 messages = [{"role": "user", "content": user_content}]
+response = client.chat.completions.create(
+    model="default-model", messages=messages, stream=args.stream, temperature=args.temp, top_p=args.top_p, tools=tools
+)
 if args.stream:
-    response = client.chat.completions.create(model="default-model", messages=messages, stream=True, tools=tools)
     for chunk in response:
         content = chunk.choices[0].delta.content
         if content is not None:
             print(content, end="", flush=True)
     print()
 else:
-    response = client.chat.completions.create(model="default-model", messages=messages, tools=tools)
     print(response.choices[0].message.content)