Deploy with Docker Compose

Prerequisites

Before installing Xpert, ensure your machine meets the following minimum system requirements:

• CPU >= 2 cores
• RAM >= 4 GiB

Clone Xpert

git clone https://github.com/xpert-ai/xpert.git
cd xpert

Start Xpert

1. Navigate to the Docker directory in the Xpert source code

cd docker

2. Copy the environment configuration file

cp env.example .env

3. Start the Docker containers

docker compose up -d

After executing the command, you should see output similar to the following, showing the status and port mappings of all containers:

 ✔ Network xpert-ai_default         Created                  0.0s 
 ✔ Container xpert-ai-db-1          Started                  0.0s 
 ✔ Container xpert-ai-redis-1       Started                  0.0s 
 ✔ Container xpert-ai-api-1         Started                  0.0s 
 ✔ Container xpert-ai-webapp-1      Started                  0.0s

Finally, check if all containers are running successfully:

docker compose ps

This includes two core services: api / webapp, and two dependency components: db / redis.

NAME                IMAGE                                       COMMAND                  SERVICE   CREATED         STATUS                   PORTS
xpert-ai-api-1      ghcr.io/xpert-ai/xpert-api:latest           "docker-entrypoint.s…"   api       4 minutes ago   Up 4 minutes             0.0.0.0:3000->3000/tcp
xpert-ai-db-1       pgvector/pgvector:pg12                      "docker-entrypoint.s…"   db        4 minutes ago   Up 4 minutes (healthy)   5432/tcp
xpert-ai-redis-1    redis/redis-stack:latest                    "sh -c 'redis-server…"   redis     4 minutes ago   Up 4 minutes             6379/tcp, 8001/tcp
xpert-ai-webapp-1   ghcr.io/xpert-ai/xpert-webapp:latest        "./entrypoint.compos…"   webapp    4 minutes ago   Up 4 minutes             0.0.0.0:80->80/tcp, 443/tcp

Enable BI Service

If you need to enable multidimensional modeling for data analysis, start the Docker containers with the bi profile:

docker compose --profile bi up -d

Start Milvus Service

If you need to use Milvus for knowledge base vector storage and retrieval:

• Modify the VECTOR_STORE configuration item in the .env file, setting it to milvus. Also, configure the Milvus account with MILVUS_USER and MILVUS_PASSWORD.
• Start the Docker containers with the milvus profile:

docker compose --profile milvus up -d

Start Crawl4AI service

If you need to use Rawl4AI To crawl web data, please add the ‘crawl4ai’ configuration when executing the Docker compose command:

docker compose --profile crawl4ai up -d

Access Xpert AI

Visit the initialization page to set up the admin account and tenant:

# Local environment
http://localhost/

# Server environment
http://your_server_ip/

Change the base URL in the .env file to the server address when deploying Xpert AI in a server environment.

API_BASE_URL=//your_server_ip:3000
CLIENT_BASE_URL=//your_server_ip

Custom Configuration

Directly edit the environment variable values in the .env file. Then, restart Xpert:

docker compose down
docker compose up -d

Upgrade

Navigate to the docker directory in the Xpert source code and execute the following commands:

cd xpert/docker
docker compose down
git pull origin main
docker compose pull
docker compose up -d

Sync Environment Variable Configuration (Important)

• If the env.example file has been updated, be sure to modify your local .env file accordingly.
• Check and adjust the configuration items in the .env file as needed to ensure they match your actual environment. You may need to add any new variables from env.example to your .env file and update any changed values.

Read More

If you have any questions, refer to the FAQ.