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

# Deploy InferenceWall in Production

> Deploy InferenceWall via PyPI, Docker, or Kubernetes Helm chart, then configure API keys, TLS, Redis sessions, and health check endpoints for production use.

InferenceWall supports three deployment paths: a Python package installed directly on the host, a Docker container, or a Kubernetes deployment via Helm. All three paths expose the same API server and SDK. Choose based on your existing infrastructure.

## Installation

<CodeGroup>
  ```bash PyPI theme={null}
  # Lite — heuristic engine only, zero ML deps
  pip install inferwall

  # Standard — adds ONNX classifier + FAISS semantic engine
  pip install inferwall[standard]

  # Full — adds LLM-judge for borderline cases
  pip install inferwall[full]
  ```

  ```bash Docker theme={null}
  docker build -t inferwall .
  docker run -p 8000:8000 \
    -e IW_API_KEY=iwk_scan_yourkey \
    -e IW_ADMIN_KEY=iwk_admin_yourkey \
    inferwall
  ```

  ```bash Kubernetes (Helm) theme={null}
  helm install inferwall ./helm/inferwall \
    --set env.IW_API_KEY=iwk_scan_yourkey \
    --set env.IW_ADMIN_KEY=iwk_admin_yourkey
  ```
</CodeGroup>

Pre-built wheels are available for Linux x86\_64, Linux aarch64, macOS arm64, and Windows x86\_64. Requires Python >= 3.10.

### Deployment profiles

| Profile      | Install                           | Engines                                | Latency     |
| ------------ | --------------------------------- | -------------------------------------- | ----------- |
| **Lite**     | `pip install inferwall`           | Heuristic (Rust)                       | \<0.3ms p99 |
| **Standard** | `pip install inferwall[standard]` | + Classifier (ONNX) + Semantic (FAISS) | \<80ms p99  |
| **Full**     | `pip install inferwall[full]`     | + LLM-Judge                            | \<2s p99    |

## Post-install setup

<Steps>
  <Step title="Generate API keys">
    ```bash theme={null}
    inferwall admin setup
    ```

    This generates a scan key (`iwk_scan_…`) and an admin key (`iwk_admin_…`) and writes them to `.env.local`.
  </Step>

  <Step title="Set environment variables">
    Export the generated keys before starting the server:

    ```bash theme={null}
    export IW_API_KEY=iwk_scan_yourkey
    export IW_ADMIN_KEY=iwk_admin_yourkey
    ```

    Or source the generated file directly:

    ```bash theme={null}
    source .env.local
    ```
  </Step>

  <Step title="Start the server">
    ```bash theme={null}
    inferwall serve

    # Or source the generated env file and serve in one command
    source .env.local && inferwall serve
    ```

    The server listens on `0.0.0.0:8000` by default.
  </Step>

  <Step title="Install ML models (Standard and Full only)">
    If you installed the `standard` or `full` profile, download the ML models:

    ```bash theme={null}
    inferwall models install --profile standard
    ```

    Models are cached in `~/.cache/inferwall/models/` and downloaded from HuggingFace (\~730 MB for Standard).
  </Step>

  <Step title="Run a health check">
    Confirm the server is up and signatures are loaded:

    ```bash theme={null}
    curl http://localhost:8000/v1/health
    ```
  </Step>
</Steps>

<Note>
  In development, you can skip API key setup entirely. Run `inferwall serve` without setting `IW_API_KEY` or `IW_ADMIN_KEY` and scan without any `Authorization` header. Dev mode is not suitable for production.
</Note>

## Environment variables

| Variable       | Description                                        | Default         |
| -------------- | -------------------------------------------------- | --------------- |
| `IW_API_KEY`   | Scan API key                                       | None (dev mode) |
| `IW_ADMIN_KEY` | Admin API key                                      | None (dev mode) |
| `IW_HOST`      | Server bind host                                   | `0.0.0.0`       |
| `IW_PORT`      | Server port                                        | `8000`          |
| `IW_TLS`       | TLS mode: `auto`, `off`, or `acme`                 | `off`           |
| `IW_PROFILE`   | Deployment profile: `lite`, `standard`, `full`     | `lite`          |
| `IW_LOG_LEVEL` | Log verbosity: `debug`, `info`, `warning`, `error` | `info`          |
| `IW_REDIS_URL` | Redis URL for distributed sessions                 | None            |

### TLS modes

| Mode   | Behavior                                                  |
| ------ | --------------------------------------------------------- |
| `off`  | Plain HTTP (default)                                      |
| `auto` | TLS using a certificate at the path provided in `IW_TLS`  |
| `acme` | Automatic certificate provisioning via ACME/Let's Encrypt |

### Redis for distributed sessions

Set `IW_REDIS_URL` to enable distributed rate limiting and session state across multiple InferenceWall instances:

```bash theme={null}
export IW_REDIS_URL=redis://redis:6379
```

When unset, InferenceWall uses in-process state, which is scoped to a single instance.

## Health check endpoints

| Endpoint               | Purpose                                            | Use in                      |
| ---------------------- | -------------------------------------------------- | --------------------------- |
| `GET /v1/health/live`  | Liveness — is the process alive?                   | Kubernetes `livenessProbe`  |
| `GET /v1/health/ready` | Readiness — can it handle requests?                | Kubernetes `readinessProbe` |
| `GET /v1/health`       | Full health with signature count and engine status | Monitoring dashboards       |

```bash theme={null}
# Liveness
curl http://localhost:8000/v1/health/live

# Readiness
curl http://localhost:8000/v1/health/ready

# Full health
curl http://localhost:8000/v1/health
```

## Further reading

<CardGroup cols={2}>
  <Card title="Environment variables reference" href="/reference/environment-variables">
    Complete list of all environment variables with types, defaults, and valid values.
  </Card>

  <Card title="Health API" href="/api/health">
    Response schemas for the liveness, readiness, and full health endpoints.
  </Card>
</CardGroup>