---
title: Ollama
description: Run models locally with Ollama and connect to Nexus — no API key required.
---

Ollama runs models locally and exposes an OpenAI-compatible API. No API key is required.

## Installation

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

## Quick Start

```go
provider := ollama.New()

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

No API key is needed. Ollama connects to `http://localhost:11434/v1` by default.

## Options

| Option | Description |
|--------|-------------|
| `ollama.WithBaseURL(url)` | Override the API base URL (default: `http://localhost:11434/v1`) |

## Capabilities

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

## Models

| Model | Context | Max Output | Price |
|-------|---------|-----------|-------|
| `llama3.1:8b` | 131K | 4,096 | Free (local) |
| `llama3.1:70b` | 131K | 4,096 | Free (local) |
| `mistral:7b` | 32,768 | 4,096 | Free (local) |
| `nomic-embed-text` | 8,192 | — | Free (local) |