MiniRobotLanguage (MRL)
AIN.
AnythingLLM AI Integration Overview
Intention
To use AnythingNLM with the API and through the web browser, it's recommended to install it via the Linux emulator Docker. You can find the instructions at this link.
https://docs.anythingllm.com/installation-docker/local-docker
If you have Docker installed and are running the AnythingNL app as a container, you can open the API documentation in your browser using this link.
http://localhost:3001/api/docs/
AnythingLLM AI Command Suite
The AIN. Command suite integrates MiniRobotLanguage with AnythingLLM, a versatile AI platform by Mintplex Labs, supporting text chat, vision queries, and AI agents.
Use subCommands to configure and interact with the API (default: http://localhost:3001/api).
Overview
AIN. provides a gateway to AnythingLLM’s features, including:
•Text Chat: Completions with LLMs like GPT-4, LLaMA, or Grok (AIN.Ask).
•Vision Queries: Image analysis with multi-modal models (AIN.AskV).
•Configuration: Set API keys (AIN.SetKey), models, and endpoints.
•RAG & Agents: Leverage document embeddings and tool-using AI agents.
Key Features
- LLMs: LLaMA (default), OpenAI (GPT-4), Google Gemini Pro, Anthropic Claude, Grok (xAI), Ollama, Mistral.
- Embedding Models: Built-in, OpenAI (text-embedding-ada-002), Cohere, Hugging Face.
- Vector Databases: LanceDB (default), Pinecone, Chroma, Qdrant.
- Defaults: Max tokens 4096, temperature 0.7; adjustable via subCommands.
Getting Started
1. Set API key with AIN.SetKey (retrieve from /api/system/api-key).
2. Optionally configure model (AIN.SetModel) or endpoint (AIN.SetChatEndpoint).
3. Use subCommands like AIN.Ask or AIN.AskV.
Example
AIN.SetKey|your_api_key_here
AIN.SetModel|openai/gpt-4
AIN.Ask|What is AI?|$$RES
DBP.Response: $$RES
SubCommands
• AIN.SetKey - Set the API key.
• AIN.SetModel - Specify the LLM (e.g., GPT-4, Claude).
• AIN.SetChatEndpoint - Set custom chat endpoint.
• AIN.Ask - Perform text chat completion.
• AIN.AskV - Perform vision-based query.
• AIN.SetMaxToken - Adjust max output tokens.
• AIN.SetTemperature - Adjust response creativity.
• AIN.SetFolder - Set working folder path.
• AIN.SetResponseFormat - Set response format.
• AIN.GetKey - Get current API key.
• AIN.GetModel - Get current model.
• AIN.GetChatEndpoint - Get current chat endpoint.
• AIN.GetMaxToken - Get max token limit.
• AIN.GetTemperature - Get temperature setting.
• AIN.GetFolder - Get working folder path.
• AIN.GetResponseFormat - Get response format.
• AIN.GetContent - Get last response content.
• AIN.GetRaw - Get raw response.
• AIN.GetFinishReason - Get finish reason.
• AIN.GetTotalTokens - Get total tokens used.
• AIN.GetWordCount - Get word count of response.
• AIN.SaveKey - Save API key to file.
• AIN.Image - Placeholder for image generation.
Remarks
- Requires AnythingLLM instance (Docker or desktop app).
- Check /api/docs for endpoint specifics.
Limitations
- Image generation not supported as of March 20, 2025.
- Multi-user mode requires cloud setup.
See also: