145 lines
3.9 KiB
Markdown
145 lines
3.9 KiB
Markdown
# Mem0 REST API Server
|
||
|
||
Mem0 provides a REST API server (written using FastAPI). Users can perform all operations through REST endpoints. The API also includes OpenAPI documentation, accessible at `/docs` when the server is running.
|
||
|
||
## Features
|
||
|
||
- **Create memories:** Create memories based on messages for a user, agent, or run.
|
||
- **Retrieve memories:** Get all memories for a given user, agent, or run.
|
||
- **Search memories:** Search stored memories based on a query.
|
||
- **Update memories:** Update an existing memory.
|
||
- **Delete memories:** Delete a specific memory or all memories for a user, agent, or run.
|
||
- **Reset memories:** Reset all memories for a user, agent, or run.
|
||
- **OpenAPI Documentation:** Accessible via `/docs` endpoint.
|
||
|
||
## Technology Stack
|
||
|
||
- **Vector Store**: ChromaDB - 高性能向量数据库
|
||
- **Graph Store**: SQLite - 轻量级图存储
|
||
- **LLM & Embedder**: Ollama - 本地大模型服务
|
||
|
||
## Quick Start
|
||
|
||
### 方式 1: Docker Compose（最简单）
|
||
|
||
```bash
|
||
# 1. 复制环境变量配置
|
||
cp .env.example .env
|
||
|
||
# 2. 编辑 .env 文件，根据需要修改配置
|
||
# vi .env 或 nano .env
|
||
|
||
# 3. 启动所有服务（包括 Ollama）
|
||
docker-compose --profile gpu up -d
|
||
|
||
# 4. 访问 API 文档
|
||
open http://localhost:8888/docs
|
||
```
|
||
|
||
**注意**: 首次启动需要下载 Ollama 模型，请等待约 60 秒。
|
||
|
||
### 方式 2: Docker 命令部署
|
||
|
||
```bash
|
||
# 1. 构建镜像
|
||
docker build -t mem0-server .
|
||
|
||
# 2. 运行容器（通过环境变量传入配置）
|
||
docker run -d \
|
||
--name mem0-server \
|
||
-p 8000:8000 \
|
||
-v chroma_data:/app/chroma_db \
|
||
-v sqlite_data:/app/graph_store.db \
|
||
-v history_data:/app/history \
|
||
--add-host=host.docker.internal:host-gateway \
|
||
-e CHROMA_DB_PATH=/app/chroma_db \
|
||
-e CHROMA_COLLECTION_NAME=memories \
|
||
-e SQLITE_DB_PATH=/app/graph_store.db \
|
||
-e OLLAMA_HOST=http://host.docker.internal:11434 \
|
||
-e OLLAMA_LLM_MODEL=llama3.2 \
|
||
-e OLLAMA_EMBEDDER_MODEL=nomic-embed-text \
|
||
-e REDIS_HOST=host.docker.internal \
|
||
-e REDIS_PORT=6379 \
|
||
mem0-server
|
||
```
|
||
|
||
## 环境变量配置
|
||
|
||
所有配置都可以通过环境变量传入：
|
||
|
||
| 变量名 | 说明 | 默认值 |
|
||
|--------|------|--------|
|
||
| `CHROMA_DB_PATH` | ChromaDB 数据路径 | `/app/chroma_db` |
|
||
| `CHROMA_COLLECTION_NAME` | ChromaDB 集合名称 | `memories` |
|
||
| `SQLITE_DB_PATH` | SQLite 数据库路径 | `/app/graph_store.db` |
|
||
| `OLLAMA_HOST` | Ollama 服务地址 | `http://localhost:11434` |
|
||
| `OLLAMA_LLM_MODEL` | LLM 模型名称 | `llama3.2` |
|
||
| `OLLAMA_EMBEDDER_MODEL` | Embedder 模型名称 | `nomic-embed-text` |
|
||
| `HISTORY_DB_PATH` | 历史数据库路径 | `/app/history/history.db` |
|
||
|
||
详细部署文档请查看 [DEPLOYMENT.md](./DEPLOYMENT.md)
|
||
|
||
## Prerequisites
|
||
|
||
Before running the server, ensure you have:
|
||
|
||
1. **Ollama installed and running** on your host machine
|
||
```bash
|
||
# Install Ollama from https://ollama.ai
|
||
ollama pull llama3.2
|
||
ollama pull nomic-embed-text
|
||
```
|
||
|
||
2. **Docker and Docker Compose** installed
|
||
|
||
## API Endpoints
|
||
|
||
- `POST /memories` - Create new memories
|
||
- `GET /memories` - Retrieve all memories
|
||
- `GET /memories/{memory_id}` - Get a specific memory
|
||
- `POST /search` - Search memories
|
||
- `PUT /memories/{memory_id}` - Update a memory
|
||
- `DELETE /memories/{memory_id}` - Delete a memory
|
||
- `DELETE /memories` - Delete all memories for an identifier
|
||
- `POST /reset` - Reset all memories
|
||
- `GET /docs` - OpenAPI documentation
|
||
|
||
# 验证
|
||
## 测试 ollama
|
||
|
||
curl http://localhost:11434/api/embeddings \
|
||
-d '{
|
||
"model":"nomic-embed-text",
|
||
"prompt":"hello world"
|
||
}'
|
||
|
||
## 测试 mem0
|
||
健康检查：
|
||
|
||
curl http://localhost:8000/health
|
||
|
||
返回：
|
||
|
||
{"status":"ok"}
|
||
|
||
创建记忆：
|
||
|
||
curl -X POST http://localhost:8000/v2/memories \
|
||
-H "Content-Type: application/json" \
|
||
-d '{
|
||
"user_id":"test",
|
||
"text":"测试用例创建"
|
||
}'
|
||
|
||
搜索记忆：
|
||
|
||
curl -X POST http://localhost:8000/v2/memories/search \
|
||
-H "Content-Type: application/json" \
|
||
-d '{
|
||
"user_id":"test",
|
||
"query":"测试"
|
||
}'
|
||
|
||
|
||
## 测试 neo4j
|