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

# Chat with Open AGIs

> Anarchy hosts a variety of LLMs that have been fine-tuned to community specifications for specific purposes.

## Browse Open Models

On the 'AGIs' tab, you will see a grid of chatbots created in the open by the community.

Simply click on the bot you want to chat with, and you will be directed to a page where you can chat with them after signing into the application via Google or Github.

<Card title="Start Chatting" icon="circle-play" href="https://chat.dev/dashboard/models">
  Browse our openly available AI models. Filter by owner, finetune, proxy, agent, and foundation model.
</Card>

<Info>Keep in mind, all chats with open chatbots are logged in public [here!](https://chat.dev/dashboard/conversation)</Info>

## Model Types

There are essentially two types of models available to chat with on Anarchy: foundation models, and finetuned models.

### Foundational Models

Foundational models are the backbone of the AI world, developed by leading researchers and companies using vast computational resources.
These models, such as GPT (OpenAI), Claude (Anthropic), Llama (Meta), Mistral, and Mixtral are trained on extensive datasets to understand and generate human-like text.
Anarchy serves these foundational models 'out of the box' in their pure form without any additional training, providing users with powerful and versatile tools straight from the source without additional customization.

Use foundational models for tasks that require broad general knowledge, natural language processing, and a high degree of flexibility. Tasks like

<CardGroup cols={2}>
  <Card title="General-purpose Queries" icon="question">
    General-purpose queries
  </Card>

  <Card title="Extensive Knowledge Retrieval" icon="book">
    Extensive knowledge retrieval
  </Card>

  <Card title="Complex Problem-Solving" icon="brain">
    Complex problem-solving
  </Card>

  <Card title="Language Understanding and Generation" icon="language">
    Language understanding and generation tasks
  </Card>
</CardGroup>

### Finetuned Models

Finetuned models are tailored versions of foundational models, enhanced through additional training to specialize in specific tasks or domains. These models benefit from the foundational models' capabilities but are refined to excel in particular areas, making them highly efficient for targeted applications.

Finetuned models are perfect for:

<CardGroup cols={2}>
  <Card title="Domain-specific Knowledge" icon="graduation-cap">
    Domain-specific knowledge and expertise
  </Card>

  <Card title="Customized Responses" icon="comments">
    Customized responses aligned with user preferences
  </Card>

  <Card title="Enhanced Performance" icon="chart-line">
    Enhanced performance in specialized tasks
  </Card>

  <Card title="Improved Accuracy" icon="bullseye">
    Improved accuracy and relevance for niche queries
  </Card>
</CardGroup>

By leveraging finetuned models, users can access AI that is not only powerful but also precisely tuned to meet specific needs, providing more relevant and effective interactions.