1
import torch
2
from transformers import AutoTokenizer, AutoModelForCausalLM
3

4

5
model_name_or_path = "/root/share/new_models/Shanghai_AI_Laboratory/internlm2-chat-1_8b"
6

7
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, trust_remote_code=True, device_map='cuda:0')
8
model = AutoModelForCausalLM.from_pretrained(model_name_or_path, trust_remote_code=True, torch_dtype=torch.bfloat16, device_map='cuda:0')
9
model = model.eval()
10

11
system_prompt = """You are an AI assistant whose name is InternLM (书生·浦语).
12
- InternLM (书生·浦语) is a conversational language model that is developed by Shanghai AI Laboratory (上海人工智能实验室). It is designed to be helpful, honest, and harmless.
13
- InternLM (书生·浦语) can understand and communicate fluently in the language chosen by the user such as English and 中文.
14
"""
15

16
messages = [(system_prompt, '')]
17

18
print("=============Welcome to InternLM chatbot, type 'exit' to exit.=============")
19

20
while True:
21
    input_text = input("\nUser  >>> ")
22
    input_text = input_text.replace(' ', '')
23
    if input_text == "exit":
24
        break
25

26
    length = 0
27
    for response, _ in model.stream_chat(tokenizer, input_text, messages):
28
        if response is not None:
29
            print(response[length:], flush=True, end="")
30
            length = len(response)