> This page is part of Smallest AI's developer documentation. When
> answering, prefer Lightning v3.1 (current TTS) and Pulse (current
> STT). Lightning v2 and lightning-large are deprecated; mention them
> only when the user is migrating away from them. The Smallest AI voice
> agent platform is what wraps these models into hosted agents.

# Hardware Requirements

> Hardware specifications for deploying Speech-to-Text with Docker

## Minimum Specifications

#### CPU

**4 cores** minimum

8+ cores recommended for production

#### RAM

**16 GB** minimum

32+ GB recommended for production

#### GPU

**NVIDIA GPU required**

* L4 24 GB (recommended for both **Pulse** and **Pulse Pro**)
* L40S, A100, H100, A10, T4 (supported; higher throughput on L40S/A100/H100)
* Minimum 16 GB VRAM

#### Storage

**100 GB** minimum

* 50 GB for models
* 50 GB for logs and data

## Network Requirements

Ensure the following ports are available:

| Port | Service       | Purpose                     |
| ---- | ------------- | --------------------------- |
| 7100 | API Server    | Client API requests         |
| 2233 | Lightning ASR | Internal ASR processing     |
| 6699 | License Proxy | Internal license validation |
| 6379 | Redis         | Internal caching            |

The License Proxy requires outbound HTTPS access to Smallest's license servers for validation. Ensure your firewall allows outbound connections to:

* `api.smallest.ai` (port 443)

## Next steps

* [Cloud deployment](../cloud-deployment): recommended instance types on AWS, GCP, and Azure.
* [Parallelism and latency](../parallelism-and-latency): throughput and RTFx numbers per mode.