alexsun/compose-anything
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: add more Agent services & easytier

Browse Source
This commit is contained in:
Summer Shen
2026-04-19 12:26:54 +08:00
parent 0e948befac
commit 0b5ba69cb0
30 changed files with 1775 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/anythingllm/.env.example
+22
View File
@@ -0,0 +1,22 @@
# Global Registry Prefix (optional)
# GLOBAL_REGISTRY=
# AnythingLLM Image Version
# No stable semantic version tags exist; 'latest' tracks the current release.
ANYTHINGLLM_VERSION=latest
# Timezone
TZ=UTC
# Host port for the AnythingLLM web UI
ANYTHINGLLM_PORT_OVERRIDE=3001
# UID/GID for file ownership inside the container
ANYTHINGLLM_UID=1000
ANYTHINGLLM_GID=1000
# Resource Limits
ANYTHINGLLM_CPU_LIMIT=2
ANYTHINGLLM_MEMORY_LIMIT=2G
ANYTHINGLLM_CPU_RESERVATION=0.5
ANYTHINGLLM_MEMORY_RESERVATION=512M
src/anythingllm/README.md
+49
View File
@@ -0,0 +1,49 @@
# AnythingLLM
[English](./README.md) | [中文](./README.zh.md)
Quick start: <https://docs.anythingllm.com>.
This service deploys AnythingLLM, an all-in-one AI application that lets you chat with documents, use multiple LLM providers, and build custom AI agents — with a full RAG pipeline built in.
## Services
- `anythingllm`: The AnythingLLM web application.
## Quick Start
```bash
docker compose up -d
```
Open `http://localhost:3001` and complete the setup wizard to connect your LLM provider.
## Configuration
All LLM providers, vector databases, and agent settings are configured through the web UI after startup. No API keys are required in `.env` unless you want to pre-seed them via environment variables.
| Variable | Description | Default |
| ----------------------------- | ----------------------------------------------- | -------- |
| `ANYTHINGLLM_VERSION` | Image version (`latest` — no stable tags exist) | `latest` |
| `TZ` | Container timezone | `UTC` |
| `ANYTHINGLLM_PORT_OVERRIDE` | Host port for the web UI | `3001` |
| `ANYTHINGLLM_UID` | UID for volume file ownership | `1000` |
| `ANYTHINGLLM_GID` | GID for volume file ownership | `1000` |
| `ANYTHINGLLM_CPU_LIMIT` | CPU limit | `2` |
| `ANYTHINGLLM_MEMORY_LIMIT` | Memory limit | `2G` |
| `ANYTHINGLLM_CPU_RESERVATION` | CPU reservation | `0.5` |
| `ANYTHINGLLM_MEMORY_LIMIT` | Memory reservation | `512M` |
## Volumes
- `anythingllm_storage`: Persists all application data, uploaded documents, embeddings, and settings.
## Ports
- **3001**: Web UI
## Notes
- The `mintplexlabs/anythingllm` image does not publish stable semantic version tags; `latest` is the only reliable tag.
- Supports OpenAI, Anthropic, Ollama, LM Studio, and many other LLM backends — all configured from the UI.
- The health check uses the `/api/ping` endpoint.
src/anythingllm/README.zh.md
+49
View File
@@ -0,0 +1,49 @@
# AnythingLLM
[English](./README.md) | [中文](./README.zh.md)
快速开始：<https://docs.anythingllm.com>。
此服务用于部署 AnythingLLM，一款集文档问答、多 LLM 提供商接入和自定义 AI Agent 于一体的全能 AI 应用，内置完整的 RAG 流水线。
## 服务
- `anythingllm`：AnythingLLM Web 应用。
## 快速开始
```bash
docker compose up -d
```
打开 `http://localhost:3001`，按照设置向导连接你的 LLM 提供商。
## 配置
所有 LLM 提供商、向量数据库和 Agent 设置均通过启动后的 Web UI 进行配置，无需在 `.env` 中预设 API Key（除非你希望通过环境变量预填充）。
| 变量 | 说明 | 默认值 |
| ----------------------------- | ----------------------------------- | -------- |
| `ANYTHINGLLM_VERSION` | 镜像版本（无语义化稳定标签，使用 `latest`） | `latest` |
| `TZ` | 容器时区 | `UTC` |
| `ANYTHINGLLM_PORT_OVERRIDE` | Web UI 的宿主机端口 | `3001` |
| `ANYTHINGLLM_UID` | 数据卷文件所有者 UID | `1000` |
| `ANYTHINGLLM_GID` | 数据卷文件所有者 GID | `1000` |
| `ANYTHINGLLM_CPU_LIMIT` | CPU 限制 | `2` |
| `ANYTHINGLLM_MEMORY_LIMIT` | 内存限制 | `2G` |
| `ANYTHINGLLM_CPU_RESERVATION` | CPU 预留 | `0.5` |
| `ANYTHINGLLM_MEMORY_LIMIT` | 内存预留 | `512M` |
## 数据卷
- `anythingllm_storage`：持久化所有应用数据、上传的文档、嵌入向量和配置。
## 端口
- **3001**：Web UI
## 说明
- `mintplexlabs/anythingllm` 镜像未发布语义化稳定标签，`latest` 是唯一可靠的标签。
- 支持 OpenAI、Anthropic、Ollama、LM Studio 等众多 LLM 后端，均可在 UI 中配置。
- 健康检查使用 `/api/ping` 端点。
src/anythingllm/docker-compose.yaml
+42
View File
@@ -0,0 +1,42 @@
x-defaults: &defaults
restart: unless-stopped
logging:
driver: json-file
options:
max-size: 100m
max-file: '3'
services:
anythingllm:
<<: *defaults
image: ${GLOBAL_REGISTRY:-}mintplexlabs/anythingllm:${ANYTHINGLLM_VERSION:-latest}
ports:
- '${ANYTHINGLLM_PORT_OVERRIDE:-3001}:3001'
volumes:
- anythingllm_storage:/app/server/storage
environment:
- TZ=${TZ:-UTC}
- STORAGE_DIR=/app/server/storage
- UID=${ANYTHINGLLM_UID:-1000}
- GID=${ANYTHINGLLM_GID:-1000}
healthcheck:
test:
- CMD
- node
- -e
- "require('http').get('http://localhost:3001/api/ping',res=>process.exit(res.statusCode===200?0:1)).on('error',()=>process.exit(1))"
interval: 30s
timeout: 10s
retries: 5
start_period: 30s
deploy:
resources:
limits:
cpus: ${ANYTHINGLLM_CPU_LIMIT:-2}
memory: ${ANYTHINGLLM_MEMORY_LIMIT:-2G}
reservations:
cpus: ${ANYTHINGLLM_CPU_RESERVATION:-0.5}
memory: ${ANYTHINGLLM_MEMORY_RESERVATION:-512M}
volumes:
anythingllm_storage:
src/easytier/.env.example
+30
View File
@@ -0,0 +1,30 @@
# EasyTier image version
EASYTIER_VERSION=v2.6.0
# Timezone
TZ=UTC
# Virtual network name (shared with all peers in the same network)
EASYTIER_NETWORK_NAME=easytier
# Virtual network secret — REQUIRED, change before deploying
# Generate a strong secret: openssl rand -hex 16
EASYTIER_NETWORK_SECRET=
# Virtual IPv4 address of this server node within the EasyTier network
EASYTIER_IPV4=10.144.144.1
# Host port for peer TCP connections
EASYTIER_TCP_PORT_OVERRIDE=11010
# Host port for peer UDP connections
EASYTIER_UDP_PORT_OVERRIDE=11010
# Host port for the management RPC portal (bound to 127.0.0.1 by default)
EASYTIER_RPC_PORT_OVERRIDE=15888
# Resource limits
EASYTIER_CPU_LIMIT=0.50
EASYTIER_MEMORY_LIMIT=128M
EASYTIER_CPU_RESERVATION=0.10
EASYTIER_MEMORY_RESERVATION=32M
src/easytier/README.md
+88
View File
@@ -0,0 +1,88 @@
# EasyTier
[English](./README.md) | [中文](./README.zh.md)
[EasyTier](https://github.com/EasyTier/EasyTier) is a mesh VPN networking tool that lets you build a private, encrypted overlay network across hosts that are behind NAT or firewalls. This stack deploys EasyTier as a **public relay server** — a stable entry point that peers can use for discovery and traffic relay when direct connections are not possible.
## Services
- `easytier`: EasyTier core node running in relay-only mode (`--no-tun`), without creating a local TUN interface.
## Ports
| Port | Protocol | Description |
| ---------- | -------- | ------------------------------------------------------------------ |
| `11010` | TCP | Peer connection listener — must be publicly reachable |
| `11010` | UDP | Peer connection listener — must be publicly reachable |
| `15888` | TCP | Management RPC portal (bound to `127.0.0.1` by default) |
## Environment Variables
| Variable | Description | Default |
| --------------------------- | --------------------------------------------------- | ---------------- |
| `EASYTIER_VERSION` | EasyTier image version | `v2.6.0` |
| `TZ` | Timezone | `UTC` |
| `EASYTIER_NETWORK_NAME` | Virtual network name shared by all peers | `easytier` |
| `EASYTIER_NETWORK_SECRET` | Network secret (password); **required** | `""` |
| `EASYTIER_IPV4` | Virtual IPv4 of this server node | `10.144.144.1` |
| `EASYTIER_TCP_PORT_OVERRIDE`| Host port for peer TCP listener | `11010` |
| `EASYTIER_UDP_PORT_OVERRIDE`| Host port for peer UDP listener | `11010` |
| `EASYTIER_RPC_PORT_OVERRIDE`| Host port for management RPC (localhost only) | `15888` |
| `EASYTIER_CPU_LIMIT` | CPU limit | `0.50` |
| `EASYTIER_MEMORY_LIMIT` | Memory limit | `128M` |
## Quick Start
1. Copy `.env.example` and set a strong network secret:
```bash
cp .env.example .env
```
Edit `.env`:
```env
EASYTIER_NETWORK_NAME=myvpn
EASYTIER_NETWORK_SECRET=<your-strong-secret>
```
Generate a secret with: `openssl rand -hex 16`
2. Start the server:
```bash
docker compose up -d
```
3. Verify the node is healthy:
```bash
docker compose exec easytier easytier-cli -p 127.0.0.1:15888 node info
```
4. On each peer machine, connect to this server:
```bash
easytier-core \
--network-name myvpn \
--network-secret <your-strong-secret> \
--peers tcp://<server-public-ip>:11010 \
--ipv4 10.144.144.2
```
## Storage
This stack does not use persistent volumes. Configuration is provided entirely via command-line flags derived from environment variables.
## Security Notes
- **`EASYTIER_NETWORK_SECRET` is required.** An empty secret leaves the network open to any peer that knows the network name. Always set a strong random value before exposing this server to the internet.
- The management RPC port (`15888`) is bound to `127.0.0.1` by default. Do not expose it publicly unless you have separate authentication in place.
- Ports `11010/tcp` and `11010/udp` must be open in your firewall / cloud security group for peers to reach this server.
- This stack runs in `--no-tun` relay mode. No kernel TUN device is created, so no elevated capabilities (`NET_ADMIN`) are required and `cap_drop: ALL` is applied.
- If you need this server node to also participate as a VPN peer (with a local virtual interface), remove `--no-tun` from `command` and add `cap_add: [NET_ADMIN]` to the service.
## Documentation
- [EasyTier GitHub](https://github.com/EasyTier/EasyTier)
- [EasyTier Documentation](https://www.easytier.top/guide/introduction.html)
src/easytier/README.zh.md
+88
View File
@@ -0,0 +1,88 @@
# EasyTier
[English](./README.md) | [中文](./README.zh.md)
[EasyTier](https://github.com/EasyTier/EasyTier) 是一款网状 VPN 组网工具，可在 NAT 或防火墙后面的主机之间构建私有加密覆盖网络。本配置将 EasyTier 部署为**公共中继服务器**——作为稳定的入口节点，供各客户端节点在无法直连时进行发现和流量中转。
## 服务
- `easytier`：以中继模式（`--no-tun`）运行的 EasyTier 核心节点，不创建本地 TUN 网络接口。
## 端口
| 端口 | 协议 | 说明 |
| ------- | ---- | ------------------------------------------------- |
| `11010` | TCP | 节点连接监听端口——需公网可达 |
| `11010` | UDP | 节点连接监听端口——需公网可达 |
| `15888` | TCP | 管理 RPC 端口（默认仅绑定 `127.0.0.1`） |
## 环境变量
| 变量名 | 描述 | 默认值 |
| ---------------------------- | --------------------------------- | -------------- |
| `EASYTIER_VERSION` | EasyTier 镜像版本 | `v2.6.0` |
| `TZ` | 时区 | `UTC` |
| `EASYTIER_NETWORK_NAME` | 所有节点共享的虚拟网络名称 | `easytier` |
| `EASYTIER_NETWORK_SECRET` | 网络密钥（密码），**必须设置** | `""` |
| `EASYTIER_IPV4` | 本服务器节点在虚拟网络中的 IPv4 | `10.144.144.1` |
| `EASYTIER_TCP_PORT_OVERRIDE` | 节点 TCP 监听端口（宿主机映射） | `11010` |
| `EASYTIER_UDP_PORT_OVERRIDE` | 节点 UDP 监听端口（宿主机映射） | `11010` |
| `EASYTIER_RPC_PORT_OVERRIDE` | 管理 RPC 端口（仅本机可访问） | `15888` |
| `EASYTIER_CPU_LIMIT` | CPU 上限 | `0.50` |
| `EASYTIER_MEMORY_LIMIT` | 内存上限 | `128M` |
## 快速开始
1. 复制 `.env.example` 并设置强网络密钥：
```bash
cp .env.example .env
```
编辑 `.env`：
```env
EASYTIER_NETWORK_NAME=myvpn
EASYTIER_NETWORK_SECRET=<你的强密钥>
```
生成随机密钥：`openssl rand -hex 16`
2. 启动服务：
```bash
docker compose up -d
```
3. 验证节点状态：
```bash
docker compose exec easytier easytier-cli -p 127.0.0.1:15888 node info
```
4. 在各客户端机器上连接到此服务器：
```bash
easytier-core \
--network-name myvpn \
--network-secret <你的强密钥> \
--peers tcp://<服务器公网 IP>:11010 \
--ipv4 10.144.144.2
```
## 数据卷
本配置不使用持久化卷，所有配置均通过环境变量转换为命令行参数传入。
## 安全说明
- **`EASYTIER_NETWORK_SECRET` 为必填项。** 若密钥为空，任何知道网络名称的节点均可接入，请务必在公网暴露前设置强密钥。
- 管理 RPC 端口（`15888`）默认仅绑定 `127.0.0.1`，请勿在无额外认证保护的情况下对外暴露。
- 防火墙及云安全组需放行 `11010/tcp` 和 `11010/udp`，客户端节点才能连接到本服务器。
- 本配置以 `--no-tun` 中继模式运行，无需创建 TUN 设备，因此无需提升内核权限（`NET_ADMIN`），已应用 `cap_drop: ALL`。
- 如需服务器节点同时作为 VPN 网络中的普通成员（拥有本地虚拟网卡），请移除 `command` 中的 `--no-tun` 参数，并在服务中添加 `cap_add: [NET_ADMIN]`。
## 文档
- [EasyTier GitHub](https://github.com/EasyTier/EasyTier)
- [EasyTier 官方文档](https://www.easytier.top/guide/introduction.html)
src/easytier/docker-compose.yaml
+46
View File
@@ -0,0 +1,46 @@
x-defaults: &defaults
restart: unless-stopped
logging:
driver: json-file
options:
max-size: 100m
max-file: '3'
services:
easytier:
<<: *defaults
image: ${GLOBAL_REGISTRY:-}easytier/easytier:${EASYTIER_VERSION:-v2.6.0}
command:
- --network-name=${EASYTIER_NETWORK_NAME:-easytier}
- --network-secret=${EASYTIER_NETWORK_SECRET:-}
- --ipv4=${EASYTIER_IPV4:-10.144.144.1}
- --listeners
- tcp://0.0.0.0:11010
- udp://0.0.0.0:11010
- --rpc-portal=0.0.0.0:15888
- --no-tun
ports:
# Peer listener ports — must be reachable from the public internet
- '${EASYTIER_TCP_PORT_OVERRIDE:-11010}:11010/tcp'
- '${EASYTIER_UDP_PORT_OVERRIDE:-11010}:11010/udp'
# Management RPC — bind to localhost only by default for security
- '127.0.0.1:${EASYTIER_RPC_PORT_OVERRIDE:-15888}:15888'
environment:
- TZ=${TZ:-UTC}
# No TUN interface in server-relay mode; no special capabilities required
cap_drop:
- ALL
deploy:
resources:
limits:
cpus: ${EASYTIER_CPU_LIMIT:-0.50}
memory: ${EASYTIER_MEMORY_LIMIT:-128M}
reservations:
cpus: ${EASYTIER_CPU_RESERVATION:-0.10}
memory: ${EASYTIER_MEMORY_RESERVATION:-32M}
healthcheck:
test: ['CMD', 'easytier-cli', '-p', '127.0.0.1:15888', 'node', 'info']
interval: 30s
timeout: 10s
retries: 3
start_period: 15s
src/letta/.env.example
+23
View File
@@ -0,0 +1,23 @@
# Global Registry Prefix (optional)
# GLOBAL_REGISTRY=
# Letta Image Version
LETTA_VERSION=0.16.7
# Timezone
TZ=UTC
# Host port for the Letta REST API server
LETTA_PORT_OVERRIDE=8283
# LLM Provider API Keys (optional; at least one is required for agent functionality)
# OPENAI_API_KEY=sk-...
# ANTHROPIC_API_KEY=sk-ant-...
# GROQ_API_KEY=gsk_...
# OLLAMA_BASE_URL=http://host.docker.internal:11434
# Resource Limits
LETTA_CPU_LIMIT=1
LETTA_MEMORY_LIMIT=1G
LETTA_CPU_RESERVATION=0.25
LETTA_MEMORY_RESERVATION=256M
src/letta/README.md
+49
View File
@@ -0,0 +1,49 @@
# Letta
[English](./README.md) | [中文](./README.zh.md)
Quick start: <https://docs.letta.com>.
This service deploys Letta (formerly MemGPT), a framework for building stateful AI agents with long-term memory, persistent state, and tool use. Letta exposes a REST API for creating and managing agents programmatically.
## Services
- `letta`: The Letta agent server.
## Quick Start
```bash
docker compose up -d
```
The Letta REST API will be available at `http://localhost:8283`. You can interact with it via the [Letta Python SDK](https://github.com/letta-ai/letta) or the [ADE web interface](https://app.letta.com).
To connect a local LLM (Ollama), set `OLLAMA_BASE_URL` in your `.env` file before starting.
## Configuration
| Variable | Description | Default |
| ---------------------- | -------------------------------------------------------- | -------- |
| `LETTA_VERSION` | Image version | `0.16.7` |
| `TZ` | Container timezone | `UTC` |
| `LETTA_PORT_OVERRIDE` | Host port for the REST API | `8283` |
| `OPENAI_API_KEY` | OpenAI API key (optional) | *(empty)*|
| `ANTHROPIC_API_KEY` | Anthropic API key (optional) | *(empty)*|
| `GROQ_API_KEY` | Groq API key (optional) | *(empty)*|
| `OLLAMA_BASE_URL` | Ollama base URL, e.g. `http://host.docker.internal:11434`| *(empty)*|
| `LETTA_CPU_LIMIT` | CPU limit | `1` |
| `LETTA_MEMORY_LIMIT` | Memory limit | `1G` |
| `LETTA_CPU_RESERVATION`| CPU reservation | `0.25` |
## Volumes
- `letta_data`: Persists agent state, memory, and configuration at `/root/.letta`.
## Ports
- **8283**: REST API
## Notes
- At least one LLM provider API key (or `OLLAMA_BASE_URL`) is required to create functioning agents.
- The health check uses the `/health` endpoint.
src/letta/README.zh.md
+49
View File
@@ -0,0 +1,49 @@
# Letta
[English](./README.md) | [中文](./README.zh.md)
快速开始：<https://docs.letta.com>。
此服务用于部署 Letta（前身为 MemGPT），一个用于构建具备长期记忆、持久状态和工具调用能力的有状态 AI Agent 框架。Letta 提供 REST API，支持以编程方式创建和管理 Agent。
## 服务
- `letta`：Letta Agent 服务器。
## 快速开始
```bash
docker compose up -d
```
Letta REST API 将在 `http://localhost:8283` 可用。你可以通过 [Letta Python SDK](https://github.com/letta-ai/letta) 或 [ADE Web 界面](https://app.letta.com) 与其交互。
如需连接本地 LLM（Ollama），请在启动前在 `.env` 文件中设置 `OLLAMA_BASE_URL`。
## 配置
| 变量 | 说明 | 默认值 |
| ---------------------- | ----------------------------------------------------------- | -------- |
| `LETTA_VERSION` | 镜像版本 | `0.16.7` |
| `TZ` | 容器时区 | `UTC` |
| `LETTA_PORT_OVERRIDE` | REST API 的宿主机端口 | `8283` |
| `OPENAI_API_KEY` | OpenAI API Key（可选） | *(空)* |
| `ANTHROPIC_API_KEY` | Anthropic API Key（可选） | *(空)* |
| `GROQ_API_KEY` | Groq API Key（可选） | *(空)* |
| `OLLAMA_BASE_URL` | Ollama 基础 URL，例如 `http://host.docker.internal:11434` | *(空)* |
| `LETTA_CPU_LIMIT` | CPU 限制 | `1` |
| `LETTA_MEMORY_LIMIT` | 内存限制 | `1G` |
| `LETTA_CPU_RESERVATION`| CPU 预留 | `0.25` |
## 数据卷
- `letta_data`：在 `/root/.letta` 持久化 Agent 状态、记忆和配置。
## 端口
- **8283**：REST API
## 说明
- 创建可用的 Agent 至少需要一个 LLM 提供商的 API Key（或 `OLLAMA_BASE_URL`）。
- 健康检查使用 `/health` 端点。
src/letta/docker-compose.yaml
+43
View File
@@ -0,0 +1,43 @@
x-defaults: &defaults
restart: unless-stopped
logging:
driver: json-file
options:
max-size: 100m
max-file: '3'
services:
letta:
<<: *defaults
image: ${GLOBAL_REGISTRY:-}letta/letta:${LETTA_VERSION:-0.16.7}
ports:
- '${LETTA_PORT_OVERRIDE:-8283}:8283'
volumes:
- letta_data:/root/.letta
environment:
- TZ=${TZ:-UTC}
- OPENAI_API_KEY=${OPENAI_API_KEY:-}
- ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY:-}
- GROQ_API_KEY=${GROQ_API_KEY:-}
- OLLAMA_BASE_URL=${OLLAMA_BASE_URL:-}
healthcheck:
test:
- CMD
- python3
- -c
- "import urllib.request; urllib.request.urlopen('http://localhost:8283/health')"
interval: 30s
timeout: 10s
retries: 5
start_period: 20s
deploy:
resources:
limits:
cpus: ${LETTA_CPU_LIMIT:-1}
memory: ${LETTA_MEMORY_LIMIT:-1G}
reservations:
cpus: ${LETTA_CPU_RESERVATION:-0.25}
memory: ${LETTA_MEMORY_RESERVATION:-256M}
volumes:
letta_data:
src/lobe-chat/.env.example
+26
View File
@@ -0,0 +1,26 @@
# Global Registry Prefix (optional)
# GLOBAL_REGISTRY=
# LobeChat Image Version
LOBE_CHAT_VERSION=1.143.3
# Timezone
TZ=UTC
# Host port for the LobeChat web UI
LOBE_CHAT_PORT_OVERRIDE=3210
# Optional access code to restrict access (leave empty to allow anonymous access)
# ACCESS_CODE=your-secret-code
# LLM Provider API Keys (at least one is required for chat to work)
# OPENAI_API_KEY=sk-...
# OPENAI_PROXY_URL=https://your-proxy/v1
# ANTHROPIC_API_KEY=sk-ant-...
# GOOGLE_API_KEY=AIza...
# Resource Limits
LOBE_CHAT_CPU_LIMIT=0.5
LOBE_CHAT_MEMORY_LIMIT=512M
LOBE_CHAT_CPU_RESERVATION=0.1
LOBE_CHAT_MEMORY_RESERVATION=128M
src/lobe-chat/README.md
+45
View File
@@ -0,0 +1,45 @@
# LobeChat
[English](./README.md) | [中文](./README.zh.md)
Quick start: <https://lobehub.com/docs>.
This service deploys LobeChat in standalone (serverless) mode — a modern, high-performance AI chat interface that supports multiple LLM providers, vision models, and plugin extensibility. No database is required; all state is stored client-side.
## Services
- `lobe-chat`: The LobeChat web application.
## Quick Start
```bash
docker compose up -d
```
Open `http://localhost:3210`. Configure your LLM API keys in the settings panel (gear icon), or set them as environment variables before starting.
## Configuration
| Variable | Description | Default |
| ---------------------- | -------------------------------------------------------- | ---------- |
| `LOBE_CHAT_VERSION` | Image version | `1.143.3` |
| `TZ` | Container timezone | `UTC` |
| `LOBE_CHAT_PORT_OVERRIDE` | Host port for the web UI | `3210` |
| `ACCESS_CODE` | Optional password to restrict access (empty = open) | *(empty)* |
| `OPENAI_API_KEY` | OpenAI API key | *(empty)* |
| `OPENAI_PROXY_URL` | Custom OpenAI-compatible API base URL | *(empty)* |
| `ANTHROPIC_API_KEY` | Anthropic API key | *(empty)* |
| `GOOGLE_API_KEY` | Google Gemini API key | *(empty)* |
| `LOBE_CHAT_CPU_LIMIT` | CPU limit | `0.5` |
| `LOBE_CHAT_MEMORY_LIMIT` | Memory limit | `512M` |
## Ports
- **3210**: Web UI
## Notes
- This is the **standalone** (client-side) mode. No PostgreSQL, S3, or auth server is needed.
- Conversation history is stored in the browser; clearing browser data loses history.
- For multi-user deployments with persistent server-side data, see the [LobeChat database mode docs](https://lobehub.com/docs/self-hosting/server-database).
- The health check uses the `/api/health` endpoint.
src/lobe-chat/README.zh.md
+45
View File
@@ -0,0 +1,45 @@
# LobeChat
[English](./README.md) | [中文](./README.zh.md)
快速开始：<https://lobehub.com/docs>。
此服务以独立（无服务器）模式部署 LobeChat，这是一款现代高性能的 AI 对话界面，支持多 LLM 提供商、视觉模型和插件扩展。无需数据库，所有状态均存储在客户端。
## 服务
- `lobe-chat`：LobeChat Web 应用。
## 快速开始
```bash
docker compose up -d
```
打开 `http://localhost:3210`。在设置面板（齿轮图标）中配置 LLM API Key，或在启动前通过环境变量设置。
## 配置
| 变量 | 说明 | 默认值 |
| ------------------------- | ------------------------------------------ | --------- |
| `LOBE_CHAT_VERSION` | 镜像版本 | `1.143.3` |
| `TZ` | 容器时区 | `UTC` |
| `LOBE_CHAT_PORT_OVERRIDE` | Web UI 的宿主机端口 | `3210` |
| `ACCESS_CODE` | 可选访问密码（空则开放访问） | *(空)* |
| `OPENAI_API_KEY` | OpenAI API Key | *(空)* |
| `OPENAI_PROXY_URL` | 自定义 OpenAI 兼容 API 基础 URL | *(空)* |
| `ANTHROPIC_API_KEY` | Anthropic API Key | *(空)* |
| `GOOGLE_API_KEY` | Google Gemini API Key | *(空)* |
| `LOBE_CHAT_CPU_LIMIT` | CPU 限制 | `0.5` |
| `LOBE_CHAT_MEMORY_LIMIT` | 内存限制 | `512M` |
## 端口
- **3210**：Web UI
## 说明
- 此为**独立**（客户端）模式，无需 PostgreSQL、S3 或认证服务器。
- 对话历史存储在浏览器中，清除浏览器数据将丢失历史记录。
- 如需多用户部署及服务端持久化数据，请参阅 [LobeChat 数据库模式文档](https://lobehub.com/docs/self-hosting/server-database)。
- 健康检查使用 `/api/health` 端点。
src/lobe-chat/docker-compose.yaml
+39
View File
@@ -0,0 +1,39 @@
x-defaults: &defaults
restart: unless-stopped
logging:
driver: json-file
options:
max-size: 100m
max-file: '3'
services:
lobe-chat:
<<: *defaults
image: ${GLOBAL_REGISTRY:-}lobehub/lobe-chat:${LOBE_CHAT_VERSION:-1.143.3}
ports:
- '${LOBE_CHAT_PORT_OVERRIDE:-3210}:3210'
environment:
- TZ=${TZ:-UTC}
- ACCESS_CODE=${ACCESS_CODE:-}
- OPENAI_API_KEY=${OPENAI_API_KEY:-}
- OPENAI_PROXY_URL=${OPENAI_PROXY_URL:-}
- ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY:-}
- GOOGLE_API_KEY=${GOOGLE_API_KEY:-}
healthcheck:
test:
- CMD
- node
- -e
- "require('http').get('http://localhost:3210/api/health',res=>process.exit(res.statusCode===200?0:1)).on('error',()=>process.exit(1))"
interval: 30s
timeout: 10s
retries: 3
start_period: 15s
deploy:
resources:
limits:
cpus: ${LOBE_CHAT_CPU_LIMIT:-0.5}
memory: ${LOBE_CHAT_MEMORY_LIMIT:-512M}
reservations:
cpus: ${LOBE_CHAT_CPU_RESERVATION:-0.1}
memory: ${LOBE_CHAT_MEMORY_RESERVATION:-128M}