---
title: Groq
description: Use Groq's ultra-fast LPU inference with Nexus.
---

Groq provides ultra-fast inference via their custom LPU (Language Processing Unit) hardware. Uses an OpenAI-compatible API.

## Installation

```go
import "github.com/xraph/nexus/providers/groq"
```

## Quick Start

```go
provider := groq.New(os.Getenv("GROQ_API_KEY"))

gw := nexus.New(
    nexus.WithProvider(provider),
)
```

## Options

| Option | Description |
|--------|-------------|
| `groq.WithBaseURL(url)` | Override the API base URL (default: `https://api.groq.com/openai/v1`) |

## Capabilities

| Capability | Supported |
|-----------|-----------|
| Chat | Yes |
| Streaming | Yes |
| Embeddings | No |
| Vision | Yes |
| Tools | Yes |
| Thinking | No |

## Models

| Model | Context | Max Output | Input Price | Output Price |
|-------|---------|-----------|-------------|-------------|
| `llama-3.3-70b-versatile` | 128K | 32,768 | $0.59/M | $0.79/M |
| `llama-3.1-8b-instant` | 131K | 8,192 | $0.05/M | $0.08/M |
| `mixtral-8x7b-32768` | 32,768 | 32,768 | $0.24/M | $0.24/M |
| `gemma2-9b-it` | 8,192 | 8,192 | $0.20/M | $0.20/M |