# LLM Sherpa

Developer APIs to Accelerate LLM Projects

Repository: https://github.com/nlmatics/llmsherpa
Canonical: https://ross.abutalabs.com/products/llm-sherpa
Homepage: https://www.nlmatics.com
Language: Jupyter Notebook
License: MIT
License Family: permissive
Last push: 2024-10-18T17:42:37+00:00

## Health v2 (maintenance only)
Score: 18/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 75
- inputs: {"age_days": 1056, "days_push": 684, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1749, forks 164 (observed 2026-08-28T04:05:31.047330+00:00)

## What it is
LLM Sherpa is a Python client library providing APIs for layout-aware PDF and document parsing to feed LLM applications, backed by the open-source nlm-ingestor service. It extracts hierarchical structure such as sections, paragraphs, tables, and lists from PDFs, DOCX, PPTX, HTML, and other formats to enable smart chunking for RAG pipelines.

## Use cases
- parse pdfs preserving layout structure for rag
- chunk pdf documents intelligently for vector indexing
- extract sections headings tables and lists from pdfs
- self-host a document ingestion server for llm apps
- convert docx pptx html files to structured text for llms
- remove repeated headers footers and watermarks from pdf text

## When to choose
- you need layout-aware pdf parsing with section hierarchy for rag pipelines
- you want to self-host a document ingestion service supporting multiple file formats
- you need smart chunking that respects paragraphs tables and nested lists

## When to avoid
- you only need plain text extraction without layout information
- you need OCR for scanned pdfs without a text layer
- you need a fully managed cloud parsing service

## Facets
- artifact type: library
- maturity: active
- function: parser, pdf, ocr, rag, llm-inference, sdk
- domain: pdf, large-language-models, developer-tools
- platform: python, self-hosted, cross-platform
- tags: pdf-parsing, layout-aware-chunking, document-ingestion, layoutpdfreader, smart-chunking, retrieval-augmented-generation, natural-language-processing, docker

## Member repositories
- nlmatics/llmsherpa (sdk) score 18
- nlmatics/nlm-ingestor (backend) score 22

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:31.047330+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 2026-08-30T03:28:45.949540+00:00, confidence not recorded.
  - readme: https://github.com/nlmatics/llmsherpa (fetched 2026-08-28T04:05:31.047330+00:00, sha d23a612211a3)
  - registry_pypi: https://pypi.org/pypi/llmsherpa/json (fetched 2026-08-29T11:06:40.762292+00:00, sha c9ca875ef475)
  - registry_pypi: https://pypi.org/pypi/nlm-ingestor/json (fetched 2026-08-29T11:06:40.764314+00:00, sha 91b2e15ff7a7)
- Data as of 2026-08-30T08:39:29.467469+00:00.