conda安装
    URL:  https://repo.anaconda.com/archive/Anaconda3-2024.06-1-Linux-x86_64.sh
    bash Anaconda3-2024.06-1-Linux-x86_64.sh   #下载好后回车yes即可
    ln -s /root/anaconda3/bin/conda  /usr/bin/conda
    conda create --name python-3.12  python=3.12.7   #创建python环境
    conda activate python-3.12  #切换环境

• 1.
• 2.
• 3.
• 4.
• 5.
• 6.

ollama下载地址：https://ollama.com/
ollama run llama3:latest

• 1.
• 2.

# 从GitHub上拉取AgentScope的源代码
git clone https://github.com/modelscope/agentscope.git
cd agentscope
# 安装需要的Python 包
pip install -e .
#创建供自己调整的目录
mkdir AI

• 1.
• 2.
• 3.
• 4.
• 5.
• 6.
• 7.

### cd AI
### vim Agentscope-test.py
# -*- coding: utf-8 -*-
import agentscope
from agentscope.agents import DialogAgent, UserAgent
from agentscope.message import Msg
# 模型配置
model_configs_name = "test-llama3"
model_configs = [
    {
        "config_name": "test-llama3",
        "model_type": "ollama_chat",
        "model_name": "llama3:latest",
        "host": "192.168.100.134:11434"   # ollama所在服务器的IP和端口，端口默认11434
    }
]
#初始化agentscope
agentscope.init(model_configs = model_configs, project = "test-1", studio_url="http://0.0.0.0:5000")

#初始化agent
dialogAgent = DialogAgent(name="小助手", model_config_name=model_configs_name, sys_prompt="你是一个AI小助手")

user = UserAgent(name="User")
x = None
while True:
    if x is not None and x.content == "exit":
        break
    x = user(x)
    x = dialogAgent(x)

• 1.
• 2.
• 3.
• 4.
• 5.
• 6.
• 7.
• 8.
• 9.
• 10.
• 11.
• 12.
• 13.
• 14.
• 15.
• 16.
• 17.
• 18.
• 19.
• 20.
• 21.
• 22.
• 23.
• 24.
• 25.
• 26.
• 27.
• 28.
• 29.

### vim start_studio.py
import agentscope


agentscope.studio.init(
  host="0.0.0.0",
  port=5000
)

• 1.
• 2.
• 3.
• 4.
• 5.
• 6.
• 7.
• 8.

#启动Agentscope Studio
nohup python start_studio.py &

#启动Agent
python Agentscope.py

• 1.
• 2.
• 3.
• 4.
• 5.