Skip to content

v0.5.1

Choose a tag to compare

View all tags
@federicopalma-pro federicopalma-pro released this 03 Aug 18:49
· 32 commits to main since this release
v0.5.1
33904e4

Progressive Summarizer RAPTOR v0.5.1

A cutting-edge text processing system that recursively analyzes documents to create hierarchical summaries at multiple levels of abstraction. RAPTOR (Recursive API for Progressive Text Organization and Refinement) preserves semantic relationships while achieving significant compression ratios, making it ideal for knowledge extraction and information retrieval tasks.

Key Features

  • Recursive progressive summarization across three hierarchical levels
  • Semantic clustering with advanced embedding models for coherent text organization
  • Intelligent token optimization with configurable thresholds
  • REST API with FastAPI for seamless integration
  • GPU acceleration for high-performance processing
  • Ollama integration for local LLM deployment

Prerequisites

  • Docker and Docker Compose (for Docker deployment)
  • NVIDIA GPU with CUDA support (recommended)
  • NVIDIA Container Toolkit (for GPU passthrough in Docker)
  • Python 3.10-3.12 (Python 3.11 recommended)
  • Ollama server for LLM integration

Changelog

  • Torch library up to 2.6
  • Docker compose fixes
Assets 2
Loading