LLPhant Documentation

Welcome to LLPhant, a comprehensive PHP Generative AI Framework designed to be as simple as possible while providing you with the tools you need to build powerful applications.

LLPhant is framework agnostic and supports multiple LLM providers including OpenAI, Anthropic, Mistral, Ollama, and services compatible with the OpenAI API.

Getting Started

• Get Started
• Use Case
• Features
  • Comparison Table of all supported Language Models
  • Supported Vector Stores
  • Supported embedding generators
• FAQ

Configuration

• Connection timeout

Core Features

• Usage
  • OpenAI
  • Atlas Cloud
  • Gemini
  • Mistral
  • Ollama
  • Anthropic
  • OpenAI compatible APIs like LocalAI
  • Chat
  • Get the token usage
  • Get the last response from OpenAI
  • Images
  • Speech to text
  • Translate audio into english
  • Customizing System Messages in Question Answering
• Tools
  • Human-in-the-Loop
• Embeddings
  • Read data
  • Document Splitter
  • Embedding Formatter
  • Embedding Generator
  • VectorStores

Advanced Features

• Question Answering
  • Multy-Query query transformation
  • Detect prompt injections
  • RetrievedDocumentsTransformer and Reranking
  • Token Usage
  • Small to Big Retrieval
  • Using tools with QuestionAnswering
  • Chat session (aka chat memory)
  • Evaluating results
• Guardrails

Related Projects

• AutoPHP

Contributors and Sponsors

• Contributors
• Sponsors

---

This is documentation for LLPhant.