Release v0.5.4

🚀 Progressive Summarizer RAPTOR - Docker Images Published

This release includes Docker images for both CPU and GPU variants of the Progressive Summarizer using RAPTOR methodology:

CPU Image

docker pull ghcr.io/smart-models/Progressive-Summarizer-RAPTOR:v0.5.4-cpu
docker pull ghcr.io/smart-models/Progressive-Summarizer-RAPTOR:latest-cpu

GPU Image (CUDA 12.1)

docker pull ghcr.io/smart-models/Progressive-Summarizer-RAPTOR:v0.5.4-gpu
docker pull ghcr.io/smart-models/Progressive-Summarizer-RAPTOR:latest-gpu

Docker Compose

# CPU deployment
cd docker
docker compose --profile cpu up -d

# GPU deployment (recommended for large documents)
cd docker
docker compose --profile gpu up -d

Quick Start

# Run CPU version
docker run -p 8080:8080 ghcr.io/smart-models/Progressive-Summarizer-RAPTOR:v0.5.4-cpu

# Run GPU version (requires nvidia-docker)
docker run --gpus all -p 8080:8080 ghcr.io/smart-models/Progressive-Summarizer-RAPTOR:v0.5.4-gpu

Features

• Progressive document summarization using RAPTOR (Recursive Abstractive Processing for Tree-Organized Retrieval)
• Hierarchical clustering and summarization for long documents
• Multi-level abstraction with configurable depth
• GPU acceleration for transformer models and embedding generation
• Support for various document formats (PDF, TXT, DOCX, etc.)
• RESTful API with streaming responses
• Configurable chunk sizes and overlap strategies
• Memory-efficient processing for large documents

RAPTOR Methodology

This implementation leverages the RAPTOR approach for:

• Building hierarchical summaries through recursive clustering
• Creating tree-structured representations of document content
• Enabling multi-scale information retrieval and summarization
• Optimizing context-aware summarization for different abstraction levels

For more information, see the README.