Metadata-Version: 2.4
Name: accuralai-rag
Version: 0.2.0
Summary: High-performance retrieval augmented generation components for the AccuralAI ecosystem.
Project-URL: homepage, https://accural-ai.web.app/
Project-URL: repository, https://github.com/AccuralAI/AccuralAI
Author: AccuralAI Maintainers
License: Apache-2.0
License-File: LICENSE
Requires-Python: >=3.10
Requires-Dist: accuralai-core
Requires-Dist: numpy>=1.23
Provides-Extra: dev
Requires-Dist: pytest; extra == 'dev'
Requires-Dist: pytest-asyncio; extra == 'dev'
Requires-Dist: pytest-mock; extra == 'dev'
Provides-Extra: faiss
Requires-Dist: faiss-cpu>=1.7.4; extra == 'faiss'
Provides-Extra: redis
Requires-Dist: redis>=5.0; extra == 'redis'
Provides-Extra: torch
Requires-Dist: torch; extra == 'torch'
Provides-Extra: transformers
Requires-Dist: sentence-transformers>=2.3.0; extra == 'transformers'
Requires-Dist: transformers>=4.36; extra == 'transformers'
Description-Content-Type: text/markdown

# accuralai-rag

High-performance Retrieval Augmented Generation (RAG) utilities built for the AccuralAI orchestration ecosystem. The package provides:

- Multi-vector embedding utilities with optional GPU + hybrid sparse support.
- Intelligent chunking and metadata extraction helpers tuned for large documents.
- Hybrid dense/sparse retrieval with reciprocal rank fusion and reranking hooks.
- Query optimizers (HyDE, decomposition, semantic augmentation) plus advanced context builders.
- UltraFastRAG pipeline that combines the components with aggressive caching and streaming-friendly hooks.

All heavy dependencies are optional. Install extras such as `torch`, `transformers`, `faiss`, or `redis` based on your production topology:

```bash
pip install -e packages/accuralai-rag[torch,transformers,faiss,redis]
```

See `accuralai_rag/pipeline.py` for the orchestrated entry point and `tests/` for lightweight examples.
