---
title: Fireworks AI
description: Use Fireworks AI for fast open-source model inference with Nexus.
---

Fireworks AI provides fast inference for open-source models with embedding support. Uses an OpenAI-compatible API.

## Installation

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

## Quick Start

```go
provider := fireworks.New(os.Getenv("FIREWORKS_API_KEY"))

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

## Options

| Option | Description |
|--------|-------------|
| `fireworks.WithBaseURL(url)` | Override the API base URL (default: `https://api.fireworks.ai/inference/v1`) |

## Capabilities

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

## Models

| Model | Context | Max Output | Input Price | Output Price |
|-------|---------|-----------|-------------|-------------|
| `accounts/fireworks/models/llama-v3p1-405b-instruct` | 131K | 4,096 | $3.00/M | $3.00/M |
| `accounts/fireworks/models/llama-v3p1-8b-instruct` | 131K | 4,096 | $0.20/M | $0.20/M |
| `accounts/fireworks/models/mixtral-8x22b-instruct` | 65,536 | 4,096 | $0.90/M | $0.90/M |