athrael-soju/Snappy

🐊 Snappy's unique approach unifies vision-language late interaction with structured OCR for region-level knowledge retrieval. Like the project? Drop a star! ⭐

51
/ 100
Established

Snappy helps you quickly find precise information within scanned documents, like PDFs or images. Instead of returning an entire page, it pinpoints and extracts only the relevant paragraphs, tables, or figures that match your query. This is ideal for researchers, analysts, or legal professionals who need to extract specific details from large collections of visual documents without sifting through irrelevant content.

Use this if you need to retrieve highly specific textual or visual sections from document images and want to understand why certain regions are considered relevant.

Not ideal if your primary need is full-text search on digital-native documents without any visual context, or if you don't require fine-grained, region-level retrieval.

document-analysis information-extraction knowledge-retrieval research-data legal-discovery
No Package No Dependents
Maintenance 10 / 25
Adoption 9 / 25
Maturity 15 / 25
Community 17 / 25

How are scores calculated?

Stars

83

Forks

15

Language

Python

License

MIT

Last pushed

Feb 09, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/vector-db/athrael-soju/Snappy"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.