> ## Documentation Index
> Fetch the complete documentation index at: https://docs.rocket.new/llms.txt
> Use this file to discover all available pages before exploring further.

# Comparison guide

> Compare AI integrations in Rocket.new: OpenAI, Anthropic, Gemini, Perplexity, and ElevenLabs for chat, content, search, analysis, and voice.

export const LlmsDirective = () => <blockquote className="llms-directive">
    For the complete documentation index, see <a href="/llms.txt">llms.txt</a>.
    For a lightweight markdown version of this page, append .md to the URL.
  </blockquote>;

<LlmsDirective />

Rocket supports five AI providers. Each has different strengths. Use this page to pick the right one for your app.

## Quick answer

* For **general-purpose chat and content generation**, use **OpenAI** (GPT models).
* For **careful, nuanced analysis and long-form writing**, use **Anthropic** (Claude models).
* For **multimodal tasks** involving images, text, or video, use **Gemini**.
* For **factual search with cited sources**, use **Perplexity**.
* For **voice generation, text-to-speech, and voice cloning**, use **ElevenLabs**.

<Note>
  Rocket uses the latest model for each provider automatically. You do not need to specify a model version when connecting.
</Note>

## Comparison table

*Scroll right to see all four providers*

| Feature                      | OpenAI                     | Anthropic                           | Gemini                             | Perplexity                    | ElevenLabs                          |
| ---------------------------- | -------------------------- | ----------------------------------- | ---------------------------------- | ----------------------------- | ----------------------------------- |
| **Best for**                 | General-purpose AI         | Nuanced analysis, long-form content | Multimodal (text + images + video) | Factual search with citations | Voice generation and text-to-speech |
| **Default model**            | Latest GPT (auto-selected) | Latest Claude (auto-selected)       | Latest Gemini (auto-selected)      | Sonar (auto-selected)         | Latest voice model (auto-selected)  |
| **Chat and conversational**  | Excellent                  | Excellent                           | Excellent                          | Good (search-focused)         | No                                  |
| **Content generation**       | Excellent                  | Excellent (especially long-form)    | Good                               | Limited                       | No                                  |
| **Code generation**          | Excellent                  | Excellent                           | Good                               | Limited                       | No                                  |
| **Image understanding**      | Yes                        | Yes                                 | Excellent (native multimodal)      | No                            | No                                  |
| **Web search and citations** | No                         | No                                  | Limited                            | Excellent                     | No                                  |
| **Voice and audio**          | No                         | No                                  | No                                 | No                            | Excellent                           |
| **Pricing**                  | Per token                  | Per token                           | Per token                          | Per token + per request       | Per character                       |
| **Connection method**        | API key                    | API key                             | API key                            | API key                       | API key                             |

<Divider />

## Detailed recommendations

### General-purpose AI chatbot or assistant

**Use OpenAI.** GPT models are the most versatile option. They handle chat, content generation, code writing, summarization, and classification well. OpenAI has the largest ecosystem and the most community examples.

```plaintext wrap theme={null}
Add an AI chat assistant to my app using OpenAI. Include a message history, streaming responses, and a text input with a send button.
```

### Careful, nuanced content or long document analysis

**Use Anthropic.** Claude excels at tasks that require careful reasoning, following complex instructions, and producing nuanced, well-structured long-form content. It handles very large context windows, making it ideal for analyzing long documents.

```plaintext wrap theme={null}
Build a document analysis tool using Anthropic Claude. Users upload a PDF, and the app generates a structured summary with key findings and recommendations.
```

### Image, video, or mixed media analysis

**Use Gemini.** Gemini is natively multimodal, meaning it understands text, images, and video in a single model. It also supports very large context windows for processing long documents or transcripts.

```plaintext wrap theme={null}
Build an image description tool using Gemini. Users upload a photo and the app generates an alt text description, color palette, and mood analysis.
```

### Factual answers with cited sources from the web

**Use Perplexity.** Perplexity is an AI-powered search engine that returns answers with inline citations. It is the best choice when accuracy and source verification matter more than creative generation.

```plaintext wrap theme={null}
Build a research assistant using Perplexity. Users type a question and the app returns a summarized answer with numbered source citations.
```

### Voice generation, text-to-speech, and audio

**Use ElevenLabs.** ElevenLabs is the best choice for converting text into natural, lifelike speech, cloning voices, and generating multilingual audio content.

```plaintext wrap theme={null}
Add a "Read aloud" button to each article that uses ElevenLabs to read the content to the user with a natural-sounding voice.
```

### Multiple AI models in the same app

**Connect multiple integrations.** You can connect OpenAI, Anthropic, Gemini, Perplexity, and ElevenLabs to the same project. Use different models for different features based on their strengths.

```plaintext wrap theme={null}
Build a writing assistant that uses OpenAI for drafting, Anthropic for editing and fact-checking, Perplexity for research with citations, and ElevenLabs for reading the final text aloud.
```

<Divider />

## Browse AI integrations

<CardGroup cols={2}>
  <Card title="OpenAI" icon="microchip-ai" href="./openai">
    GPT models for chat, content generation, and code.
  </Card>

  <Card title="Anthropic" icon="brain" href="./anthropic">
    Claude models for nuanced analysis and long-form content.
  </Card>

  <Card title="Gemini" icon="stars" href="./gemini">
    Google's multimodal AI for text, images, and video.
  </Card>

  <Card title="Perplexity" icon="sparkles" href="./perplexity">
    AI-powered search with cited, fact-based answers.
  </Card>

  <Card title="ElevenLabs" icon="waveform-lines" href="./elevenlabs">
    Voice generation, text-to-speech, and voice cloning.
  </Card>
</CardGroup>