书生·浦语大模型实战营-8G 显存玩转书生大模型 Demo

1. 任务内容

• 使用 Cli Demo 完成 InternLM2-Chat-1.8B 模型的部署，并生成 300 字小故事。
• 使用 LMDeploy 完成 InternLM-XComposer2-VL-1.8B 的部署，并完成一次图文理解对话。
• 使用 LMDeploy 完成 InternVL2-2B 的部署，并完成一次图文理解对话。

2.task1:InternLM2-Chat-1.8B 模型的部署

其代码为：

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM


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

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

system_prompt = """You are an AI assistant whose name is InternLM (书生·浦语).
- 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.
- InternLM (书生·浦语) can understand and communicate fluently in the language chosen by the user such as English and 中文.
"""

messages = [(system_prompt, '')]

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

while True:
    input_text = input("\nUser  >>> ")
    input_text = input_text.replace(' ', '')
    if input_text == "exit":
        break

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

实际测试结果如下图：在10%性能的A100上占用显存4876MB，推理时速度较慢，肉眼可见的逐字输出，并且输出的结论幻觉严重。

在streamlit上部署与之类似：

设置端口映射后在本地浏览器上打开，结果如下，效果依旧不好：

3.task2:LMDeploy部署InternLM-XComposer2-VL-1.8B

输入命令：

lmdeploy serve gradio /share/new_models/Shanghai_AI_Laboratory/internlm-xcomposer2-vl-1_8b --cache-max-entry-count 0.1

4.task3: LMDeploy部署InternVL2-2B 

输入命令：

lmdeploy serve gradio /share/new_models/OpenGVLab/InternVL2-2B --cache-max-entry-count 0.1