Skip to main content

AnythingLLM & APIpie Configuration Guide

APIpieAnythingLLM

This guide will walk you through the process of integrating AnythingLLM with APIpie to leverage the power of multiple AI models and enhance your chatbot experience.

Steps​

1. Create an Account​

  • Link: Register here
  • Follow the link and fill out the form to create your account.

2. Add Credit​

  • Link: Add Credit
  • Access the subscription section after logging in to add credits to your account.

3. Generate an API Key​

  • Link: Generate API Key
  • Navigate to the API keys section and create a new key. This key is necessary for API requests.

4. Install AnythingLLM​

5. Access the AnythingLLM Settings​

  • Configure the LLM Preference either on initial setup or by changing the settings after configured.
  • Find your prefered LLM on the APIpie.ai Dashboard and take note of the Name, MAxTokens, and MaxResponse.
  • Set the "Base URL" to https://apipie.ai/v1/
  • Paste your generated API key into the designated field.
  • Configure the appropite model name.
  • Set the "Token Context Window" to the "MaxResponse".
  • Set the "Max Tokens" and save the settings.

Initial Setup:​

Initial Setup

Changing Settings:​

Settings

6. Set the AnythingLLM Agent LLM​

Agent Setup:​

  • From the main screen select the settings option on the configured workplace
  • Change to the "Agent Configuration" tab.
  • Select "Local AI" as the provider
Agent Setup

Agent AI Settings:​

  • Set the "Local AI Base URL" to https://apipie.ai/v1

  • Enter your generated API key ( if you want separate activity statements generate a new key identifying the agent)

  • Save Settings

    Agent AI

  • Select the desired Agent AI model.

7. Start Chatting​

  • With the integration complete, you can now enjoy AI-powered conversations in AnythingLLM.
Desktop

Caveats​

  • Currently AnythingLLM is yet support APIpie TTS, STT, or Embedding.
  • If this changes or if assistance is required to onboard more APIpie features please reach out on our Discord

Additional Resources​

  • AnythingLLM Website: Explore more features and capabilities of AnythingLLM. Visit the website
  • AnythingLLM GitHub: Access the source code and contribute to the AnythingLLM project. Visit GitHub
  • AnythingLLM Documentation: Dive deeper into AnythingLLM's functionalities and configurations. Access the docs
  • Mintplex Labs: Explore Mintplex Labs the team that brings us AnythingLLM Mintplex Labs

Connect with AnythingLLM​

Additional Support​

If you encounter any issues during the integration process, please reach out to us on Discord for assistance.