> If you encounter slow image pulling speeds, you can modify the `image` field under the corresponding service in the `docker-compose.yml` file to use an alternative image source.
```
docker pull ghcr.io/666ghj/bettafish:latest
```
> ⚠️ **Note:** If the pull speed is slow, you can try replacing the registry address `ghcr.io` with `ghcr.nju.edu.cn` for faster download:
After pulling the image, you need to use the provided `docker-compose.yml` file to start the service.
#### A. Modifying `docker-compose.yml`
Open the `docker-compose.yml` file in your project and modify the `image` field under the `bettafish` service. Replace its value with the **full name of the image** you just pulled.
**Example:**
Change the original configuration:
YAML
```
services:
bettafish:
build:
context: .
dockerfile: Dockerfile
image: bettafish:latest # <-- Replace this image name with the one you pulled
container_name: bettafish
# ... other configurations
```
To (using `ghcr.io/666ghj/bettafish:latest` as an example:
YAML
```
services:
bettafish:
image: ghcr.io/666ghj/bettafish:latest # <-- Replace with the image name you pulled
container_name: bettafish
# ... other configurations
```
#### B. Starting the Service
In the directory containing the modified `docker-compose.yml` file, execute the following command to start the service:
Bash
```
docker compose up -d
```
### 3. ⚙️ System Configuration (Database and Models)
After the service starts, you need to configure the system.
#### A. Database Configuration
Please configure the database connection with the following parameters:
| **Others** | **Keep Default** | Please keep other parameters, such as database connection pool settings, at their default values. |
### Large Language Model (LLM) Configuration
After completing the database configuration, proceed to configure all the Large Language Model related parameters that you intend to use.
After completing the database configuration, please proceed to configure **all Large Language Model related parameters** to ensure the system can connect to your chosen LLM service.
Once the configuration is complete, the system should start and run normally.
Upon completing and saving all the configurations above, the system should be ready to run normally.