README.md
ContextGem: Effortless LLM extraction from documents
| Package |     |
| Quality |      |
| Tools |         |
| Docs |     |
| Community | [ ](CODE_ |
Explore / contextgem
shcherbak-ai
contextgem
Declarative LLM extraction from documents: describe Aspects and Concepts in plain language, get structured values back with paragraph- or sentence-level references and built-in justifications.
1,990 183 Python Apache-2.0updated 11 days ago
Curator's take
Its thesis is explicitly anti-RAG: stop chunking and retrieving, put the whole document in a long context window and extract in place. You declare Aspects — the sections or themes worth pulling — and Concepts — the entities, booleans, numbers, ratings and JSON objects inside them — and the framework writes the prompts, builds the validation models, and returns every item with reference_paragraphs, reference_sentences and a justification. That provenance is the point: it makes contract and filing review defensible rather than merely plausible. Two limits it states about itself: no cross-document querying or corpus-wide retrieval, so reach for a real RAG framework there; and its detailed instruction style wants a model at gpt-4o-mini level or better, so 8B local models will struggle even though LiteLLM happily points it at Ollama.
Mapped by ShipWithAI editors · links verified
Continue your stack
What teams reach for next — and why each earns a place beside contextgem. Ranked by curator confidence.
Pairs well with
marker
Datalab's 39k-star PDF-to-Markdown/JSON converter: a layout pipeline plus an optional LLM pass for tables, forms and equations, with chunk output and form-value extraction built in.
Why it fitsmarker turns the document into Markdown; contextgem then pulls declared Aspects and Concepts out of that text with sentence-level references — conversion then extraction.
anydoc
Rust library converting Word, PowerPoint, Excel, OpenDocument, RTF, EPUB, CSV and PDF to clean Markdown in single-digit milliseconds — Node/Python/WASM bindings, CLI, and an agent skill.
Why it fitsContextGem starts from text, not scans. anydoc is the Rust converter that turns Word, PowerPoint, Excel, RTF, EPUB and PDF into clean Markdown — the step you run immediately before building the Document object.
data-formulator
Microsoft's AI data-visualization workbench: connect files, DBs or Databricks, ask in plain language, and agents write the transforms behind 30+ chart types you branch and restyle.
Why it fitsDocument to decision in two hops: contextgem pulls declared Aspects and Concepts out of contracts and reports into typed values, data-formulator then explores that table in plain language and charts it.
Alternative to See all + compare →
unstract
LLM-driven platform turning unstructured documents into structured data: a no-code Prompt Studio to define extractions, then deploy as APIs or ETL pipelines. Self-hosted, AGPL + enterprise.
Why it fitsBoth turn documents into structured data with LLMs. Unstract is a platform — no-code Prompt Studio, deployable APIs and ETL around it; ContextGem is a library you compose in code, and it goes further on provenance: every value carries a paragraph- or sentence-level reference and a justification.
PageIndex
Vectorless, reasoning-based RAG — builds a hierarchical tree index from long documents so an LLM retrieves by relevance instead of similarity. No chunking, no embeddings, no vector DB.
Why it fitsBoth reject chunk-and-embed for long documents, then diverge on what replaces it. PageIndex builds a hierarchical tree the model navigates to reach the right section; ContextGem skips retrieval entirely and sends the whole document into a long context window. Navigate versus extract in place.
docling-graph
Documents to validated knowledge graphs: Docling parses, an LLM or VLM fills Pydantic schemas, and you get a directed NetworkX graph with provenance, Cypher/CSV export and HTML views.
Why it fitsBoth fill typed schemas from documents with an LLM. docling-graph aims outward at a validated knowledge graph — entities and relations across a corpus; ContextGem deliberately stays inside one document, tying each extracted value back to the sentences that support it.