> 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.

# Models overview

> Overview of the Smallest AI models: Lightning text to speech, Pulse speech to text, Hydra speech to speech, and Electron LLM, with links to each model card.

One API key, one base URL, four things you can do with speech and language.

#### [Text to Speech](/models/text-to-speech/overview)

Stream natural speech with Lightning at 44.1 kHz and 200 ms to first byte.

Lightning v3.1 Pro

Lightning v3.1

#### [Speech to Text](/models/speech-to-text/overview)

Transcribe live or recorded audio with Pulse, 21 streaming languages.

Pulse

Pulse Pro

#### [Speech to Speech](/models/speech-to-speech/overview)

Talk to Hydra over one WebSocket, audio in and audio out.

Hydra

#### [LLM](/models/llm/overview)

Run Electron, an OpenAI-compatible model built for voice agents.

Electron

## The models

### Text to speech

#### [Lightning v3.1 Pro](/model-cards/text-to-speech/lightning-v-3-1-pro)

Premium voices, 31 languages.

#### [Lightning v3.1](/model-cards/text-to-speech/lightning-v-3-1)

217 voices, cloning, 12 languages.

### Speech to text

#### [Pulse](/model-cards/speech-to-text/pulse)

Streaming and batch, 21 streaming languages.

#### [Pulse Pro](/model-cards/speech-to-text/pulse-pro)

English batch, leaderboard accuracy.

### Speech to speech

#### [Hydra](/model-cards/speech-to-speech/hydra)

Full-duplex speech to speech.

### LLM

#### [Electron](/model-cards/llm/electron)

Chat completions, 70 languages.

Numbers side by side: [Compare models](/model-cards/compare).

## Start building

#### [Getting started](/models/getting-started)

Create a key and make your first call in a minute.

#### [API reference](/api-reference/models)

Every endpoint with schemas and copyable snippets.

Plans and per-model rates are on the [pricing page](https://smallest.ai/pricing).

> Four capabilities, six models. Pick one and go.