# agent_m

**Repository Path**: cdh171688/agent_m

## Basic Information

- **Project Name**: agent_m
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-05-06
- **Last Updated**: 2026-05-06

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Medical Triage Agent（全本地离线版）

## 1. 创建 Conda 环境
```bash
conda create -n med-agent python=3.10 -y
conda activate med-agent
pip install -r requirements.txt
```

## 2. 下载本地模型（ModelScope）
```bash
modelscope download --model Qwen/Qwen2.5-7B-Instruct \
  --local_dir /root/autodl-tmp/models/Qwen2.5-7B-Instruct

modelscope download --model sentence-transformers/all-MiniLM-L6-v2 \
  --local_dir /root/autodl-tmp/models/all-MiniLM-L6-v2
```

## 3. 启动本地 vLLM（OpenAI 兼容）
```bash
python -m vllm.entrypoints.openai.api_server \
  --model /root/autodl-tmp/models/Qwen2.5-7B-Instruct \
  --served-model-name Qwen2.5-7B-Instruct \
  --host 0.0.0.0 \
  --port 8000
```

## 4. 启动 FastAPI（离线模式）
```bash
cp .env.example .env
bash run_api.sh
```

## 5. 测试请求
```bash
curl -X POST "http://127.0.0.1:9000/chat" \
  -H "Content-Type: application/json" \
  -d '{
    "message": "I have had fever for two days, 39C, and cough.",
    "user_profile": {
      "age": 32,
      "gender": "male",
      "height_cm": 175,
      "weight_kg": 70,
      "medication_history": ["aspirin", "warfarin"]
    }
  }'
```

## 6. 网页访问
浏览器打开：

```text
http://127.0.0.1:9000/
```

页面会直接调用 `/chat` 接口，支持多轮会话（`session_id` 保持）。