带有存储的 Agent

代码

cookbook/models/vllm/storage.py

from agno.agent import Agent
from agno.models.vllm import vLLM
from agno.storage.postgres import PostgresStorage
from agno.tools.duckduckgo import DuckDuckGoTools

DB_URL = "postgresql+psycopg://ai:ai@localhost:5532/ai"

agent = Agent(
    model=vLLM(id="Qwen/Qwen2.5-7B-Instruct"),
    storage=PostgresStorage(table_name="agent_sessions", db_url=DB_URL),
    tools=[DuckDuckGoTools()],
    add_history_to_messages=True,
)

agent.print_response("How many people live in Canada?")
agent.print_response("What is their national anthem called?")

确保 Postgres 数据库正在运行。

用法

1. 创建虚拟环境

   打开 Terminal 并创建一个 python 虚拟环境。

   Mac

   python3 -m venv .venv
   source .venv/bin/activate
   
2. 安装库

   pip install -U agno openai vllm sqlalchemy psycopg[binary] duckduckgo-search
   
3. 启动 Postgres 数据库

   ./cookbook/scripts/run_pgvector.sh
   
4. 启动 vLLM 服务器

   vllm serve Qwen/Qwen2.5-7B-Instruct \
       --enable-auto-tool-choice \
       --tool-call-parser hermes \
       --dtype float16 \
       --max-model-len 8192 \
       --gpu-memory-utilization 0.9
   
5. 运行 Agent

   python cookbook/models/vllm/storage.py