# D-Star-AI/dsRAG

High-performance retrieval engine for unstructured data

Repository: https://github.com/D-Star-AI/dsRAG
Canonical: https://ross.abutalabs.com/products/dsrag
Language: Python
License: MIT
License Family: permissive
Last push: 2025-11-10T23:32:19+00:00

## Health v2 (maintenance only)
Score: 48/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 51, release rhythm 35, longevity 62
- inputs: {"age_days": 868, "days_push": 296, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1589, forks 131 (observed 2026-08-28T04:05:08.155198+00:00)

## What it is
dsRAG is a high-performance retrieval engine for unstructured data, implemented as a Python library for building RAG pipelines. It improves retrieval accuracy over vanilla RAG using semantic sectioning, AutoContext contextual chunk headers, and Relevant Segment Extraction.

## Use cases
- build a RAG pipeline over dense documents like financial reports
- answer questions over legal contracts with high accuracy
- retrieve relevant passages from academic papers
- improve retrieval quality over vanilla RAG baselines
- index and query unstructured text with contextual chunk embeddings
- run question answering on FinanceBench-style benchmarks

## When to choose
- you need state-of-the-art retrieval accuracy on dense, complex documents
- vanilla RAG chunking underperforms on your corpus
- you want a Python library with MIT license to embed in your own app

## When to avoid
- you need a fully managed turnkey RAG service with UI
- your data is already structured and simple keyword search suffices
- you cannot afford LLM calls for sectioning and context generation

## Facets
- artifact type: library
- maturity: active
- function: rag, search-engine, llm-inference, nlp, machine-learning
- domain: large-language-models, artificial-intelligence
- platform: python
- tags: retrieval-engine, semantic-sectioning, autocontext, relevant-segment-extraction, unstructured-data, rag-pipeline, retrieval-augmented-generation, natural-language-processing

## Member repositories
- D-Star-AI/dsRAG (main) score 48

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:08.155198+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:54:49.416406+00:00, confidence not recorded.
  - readme: https://github.com/D-Star-AI/dsRAG (fetched 2026-08-28T04:05:08.155198+00:00, sha ddb9bd219618)
  - registry_pypi: https://pypi.org/pypi/dsrag/json (fetched 2026-08-29T11:25:31.570272+00:00, sha 686c73031c64)
- Data as of 2026-08-30T08:39:29.467469+00:00.