> ## Documentation Index
> Fetch the complete documentation index at: https://mintlify.com/ItzCrazyKns/Perplexica/llms.txt
> Use this file to discover all available pages before exploring further.

# Environment variables

> Complete reference for environment variables in Perplexica

Perplexica can be configured using environment variables, which is especially useful for Docker deployments and automation.

## How environment variables work

<Note>
  Environment variables are processed on container startup. If a required environment variable is set (with valid configuration), the corresponding provider is automatically configured.
</Note>

Environment variables:

* Are applied when Perplexica starts
* Create provider configurations automatically
* Are merged with existing configuration
* Take precedence on first initialization

## Provider configuration

Configure LLM providers using environment variables.

### Ollama

<ParamField path="OLLAMA_BASE_URL" type="string">
  Base URL for Ollama server

  **Required:** Yes (to auto-configure Ollama)

  **Docker examples:**

  * Windows/Mac: `http://host.docker.internal:11434`
  * Linux: `http://<host-ip>:11434`

  **Non-Docker:** `http://localhost:11434`
</ParamField>

```bash theme={null}
docker run -d -p 3000:3000 \
  -e OLLAMA_BASE_URL="http://host.docker.internal:11434" \
  -v perplexica-data:/home/perplexica/data \
  --name perplexica \
  itzcrazykns1337/perplexica:latest
```

### OpenAI

<ParamField path="OPENAI_API_KEY" type="string">
  Your OpenAI API key

  **Required:** Yes (to auto-configure OpenAI)
</ParamField>

<ParamField path="OPENAI_BASE_URL" type="string" default="https://api.openai.com/v1">
  Base URL for OpenAI API

  **Required:** Yes (to auto-configure OpenAI)

  **Default:** `https://api.openai.com/v1`

  For OpenAI-compatible APIs, use your custom URL
</ParamField>

```bash theme={null}
docker run -d -p 3000:3000 \
  -e OPENAI_API_KEY="sk-..." \
  -e OPENAI_BASE_URL="https://api.openai.com/v1" \
  -v perplexica-data:/home/perplexica/data \
  --name perplexica \
  itzcrazykns1337/perplexica:latest
```

### Anthropic

<ParamField path="ANTHROPIC_API_KEY" type="string">
  Your Anthropic API key

  **Required:** Yes (to auto-configure Anthropic)
</ParamField>

```bash theme={null}
docker run -d -p 3000:3000 \
  -e ANTHROPIC_API_KEY="sk-ant-..." \
  -v perplexica-data:/home/perplexica/data \
  --name perplexica \
  itzcrazykns1337/perplexica:latest
```

### Gemini

<ParamField path="GEMINI_API_KEY" type="string">
  Your Google AI API key for Gemini

  **Required:** Yes (to auto-configure Gemini)
</ParamField>

```bash theme={null}
docker run -d -p 3000:3000 \
  -e GEMINI_API_KEY="AIza..." \
  -v perplexica-data:/home/perplexica/data \
  --name perplexica \
  itzcrazykns1337/perplexica:latest
```

### Groq

<ParamField path="GROQ_API_KEY" type="string">
  Your Groq API key

  **Required:** Yes (to auto-configure Groq)
</ParamField>

```bash theme={null}
docker run -d -p 3000:3000 \
  -e GROQ_API_KEY="gsk_..." \
  -v perplexica-data:/home/perplexica/data \
  --name perplexica \
  itzcrazykns1337/perplexica:latest
```

### LM Studio

<ParamField path="LM_STUDIO_BASE_URL" type="string">
  Base URL for LM Studio server

  **Required:** Yes (to auto-configure LM Studio)

  **Default:** `http://localhost:1234`

  The `/v1` suffix is added automatically
</ParamField>

```bash theme={null}
docker run -d -p 3000:3000 \
  -e LM_STUDIO_BASE_URL="http://host.docker.internal:1234" \
  -v perplexica-data:/home/perplexica/data \
  --name perplexica \
  itzcrazykns1337/perplexica:latest
```

### Lemonade

<ParamField path="LEMONADE_BASE_URL" type="string">
  Base URL for Lemonade server

  **Required:** Yes (to auto-configure Lemonade)

  **Example:** `https://api.lemonade.ai/v1`
</ParamField>

<ParamField path="LEMONADE_API_KEY" type="string">
  Your Lemonade API key

  **Required:** No (optional)
</ParamField>

```bash theme={null}
docker run -d -p 3000:3000 \
  -e LEMONADE_BASE_URL="http://host.docker.internal:8000" \
  -e LEMONADE_API_KEY="your-key" \
  -v perplexica-data:/home/perplexica/data \
  --name perplexica \
  itzcrazykns1337/perplexica:latest
```

## Search configuration

### SearxNG

<ParamField path="SEARXNG_API_URL" type="string" default="http://localhost:8080">
  URL for your SearxNG instance

  **Default:** `http://localhost:8080` (bundled SearxNG)

  Override this when using an external SearxNG instance
</ParamField>

```bash theme={null}
# Using external SearxNG with slim image
docker run -d -p 3000:3000 \
  -e SEARXNG_API_URL="http://my-searxng:8080" \
  -v perplexica-data:/home/perplexica/data \
  --name perplexica \
  itzcrazykns1337/perplexica:slim-latest
```

<Note>
  The bundled Perplexica image automatically sets `SEARXNG_API_URL=http://localhost:8080` for the included SearxNG instance.
</Note>

## System configuration

### Data directory

<ParamField path="DATA_DIR" type="string" default="/home/perplexica/data">
  Directory where Perplexica stores configuration and data

  **Default:** `/home/perplexica/data`

  **Note:** This is automatically set in Docker images
</ParamField>

<Warning>
  Changing `DATA_DIR` is not recommended unless you have a specific use case. Ensure you mount a volume to the new location.
</Warning>

### Docker detection

<ParamField path="DOCKER" type="boolean">
  Indicates whether Perplexica is running in Docker

  **Set by:** Docker images automatically

  **Effect:** Changes default URLs (e.g., Ollama URL suggestions)
</ParamField>

## Multiple providers

You can configure multiple providers simultaneously:

```bash theme={null}
docker run -d -p 3000:3000 \
  -e OLLAMA_BASE_URL="http://host.docker.internal:11434" \
  -e OPENAI_API_KEY="sk-..." \
  -e OPENAI_BASE_URL="https://api.openai.com/v1" \
  -e ANTHROPIC_API_KEY="sk-ant-..." \
  -e GEMINI_API_KEY="AIza..." \
  -e GROQ_API_KEY="gsk_..." \
  -v perplexica-data:/home/perplexica/data \
  --name perplexica \
  itzcrazykns1337/perplexica:latest
```

<Tip>
  Configuring multiple providers gives you flexibility to choose different models for different tasks.
</Tip>

## Docker Compose example

Using environment variables with Docker Compose:

```yaml theme={null}
version: '3.8'

services:
  perplexica:
    image: itzcrazykns1337/perplexica:latest
    ports:
      - "3000:3000"
    environment:
      - OLLAMA_BASE_URL=http://host.docker.internal:11434
      - OPENAI_API_KEY=sk-your-key
      - OPENAI_BASE_URL=https://api.openai.com/v1
      - ANTHROPIC_API_KEY=sk-ant-your-key
      - SEARXNG_API_URL=http://localhost:8080
    volumes:
      - perplexica-data:/home/perplexica/data

volumes:
  perplexica-data:
```

### Using .env file

Create a `.env` file:

```bash theme={null}
# .env
OLLAMA_BASE_URL=http://host.docker.internal:11434
OPENAI_API_KEY=sk-your-key
OPENAI_BASE_URL=https://api.openai.com/v1
ANTHROPIC_API_KEY=sk-ant-your-key
GEMINI_API_KEY=AIza-your-key
GROQ_API_KEY=gsk-your-key
SEARXNG_API_URL=http://localhost:8080
```

Then use with Docker Compose:

```yaml theme={null}
version: '3.8'

services:
  perplexica:
    image: itzcrazykns1337/perplexica:latest
    ports:
      - "3000:3000"
    env_file:
      - .env
    volumes:
      - perplexica-data:/home/perplexica/data

volumes:
  perplexica-data:
```

<Warning>
  Never commit `.env` files containing API keys to version control. Add `.env` to your `.gitignore`.
</Warning>

## Non-Docker usage

For non-Docker installations, set environment variables before starting:

### Linux/Mac

```bash theme={null}
export OLLAMA_BASE_URL="http://localhost:11434"
export OPENAI_API_KEY="sk-your-key"
export OPENAI_BASE_URL="https://api.openai.com/v1"
export SEARXNG_API_URL="http://localhost:8080"

npm run start
```

### Windows (PowerShell)

```powershell theme={null}
$env:OLLAMA_BASE_URL="http://localhost:11434"
$env:OPENAI_API_KEY="sk-your-key"
$env:OPENAI_BASE_URL="https://api.openai.com/v1"
$env:SEARXNG_API_URL="http://localhost:8080"

npm run start
```

### Windows (Command Prompt)

```cmd theme={null}
set OLLAMA_BASE_URL=http://localhost:11434
set OPENAI_API_KEY=sk-your-key
set OPENAI_BASE_URL=https://api.openai.com/v1
set SEARXNG_API_URL=http://localhost:8080

npm run start
```

## Validation

Perplexica validates environment variables on startup:

* **Missing required fields:** Provider is not auto-configured
* **Invalid values:** Error logged, provider skipped
* **Connection test:** Providers are tested when configured

Check logs for validation messages:

```bash theme={null}
docker logs perplexica
```

<Accordion title="Example validation messages">
  ```
  Failed to initialize provider. Type: openai, ID: abc-123, Config: {...}, Error: Invalid config provided. API key and base URL must be provided
  ```

  ```
  Failed to get model list. Type: ollama, ID: def-456, Error: Error connecting to Ollama API. Please ensure the base URL is correct and the Ollama server is running.
  ```
</Accordion>

## Environment variable precedence

Configuration priority (highest to lowest):

1. **Existing configuration:** Settings already saved in `config.json`
2. **Environment variables:** Applied on startup for new providers
3. **Defaults:** Built-in defaults (e.g., OpenAI base URL)

<Note>
  If a provider with the same configuration already exists (matched by hash), environment variables won't create a duplicate.
</Note>

## Security considerations

<Warning>
  **Never expose API keys in:**

  * Public repositories
  * Container images
  * Logs or error messages
  * Unencrypted storage
</Warning>

Best practices:

* Use Docker secrets or orchestration tools (Kubernetes secrets, etc.)
* Rotate API keys regularly
* Use read-only API keys when possible
* Restrict network access to Perplexica
* Use HTTPS for external API endpoints

<Tip>
  For production deployments, consider using secret management tools like:

  * Docker Swarm secrets
  * Kubernetes secrets
  * HashiCorp Vault
  * AWS Secrets Manager
  * Azure Key Vault
</Tip>

## Troubleshooting

### Provider not auto-configured

<Accordion title="Check required variables are set">
  Each provider requires specific environment variables:

  * **Ollama:** `OLLAMA_BASE_URL`
  * **OpenAI:** `OPENAI_API_KEY` AND `OPENAI_BASE_URL`
  * **Anthropic:** `ANTHROPIC_API_KEY`
  * **Gemini:** `GEMINI_API_KEY`
  * **Groq:** `GROQ_API_KEY`
  * **LM Studio:** `LM_STUDIO_BASE_URL`
  * **Lemonade:** `LEMONADE_BASE_URL`

  Missing any required variable prevents auto-configuration.
</Accordion>

<Accordion title="Verify environment variables are passed">
  Check that Docker receives the variables:

  ```bash theme={null}
  docker exec perplexica env | grep -E 'OLLAMA|OPENAI|ANTHROPIC|GEMINI|GROQ|SEARXNG'
  ```
</Accordion>

<Accordion title="Check logs for errors">
  View startup logs:

  ```bash theme={null}
  docker logs perplexica 2>&1 | grep -i error
  ```
</Accordion>

### Values not updating

<Note>
  Environment variables are only processed on first initialization. If configuration already exists, environment variables won't override it.
</Note>

To force reconfiguration:

1. Remove the existing provider in the settings UI
2. Restart Perplexica
3. Environment variables will create a new provider entry

Or delete the configuration file (⚠️ removes all settings):

```bash theme={null}
docker exec perplexica rm /home/perplexica/data/config.json
docker restart perplexica
```

## Next steps

* [Configure providers](/configuration/models-and-providers)
* [Set up SearxNG](/configuration/searxng)
* [Application settings](/configuration/settings)