Cerno-AI/Cerno-Insight

High-performance RAG system for intelligent document Q&A with hybrid retrieval, GPU acceleration, and citation-backed answers. Upload docs, ask questions, get precise responses.

30
/ 100
Emerging

Implements intelligent document triage with four specialized processing modes (Direct, RAG Pipeline, Vision, Raw Text) that automatically route queries based on document size and type for optimal latency. Features hybrid retrieval combining BM25 keyword search with GPU-accelerated FAISS vector similarity, reciprocal rank fusion, and CrossEncoder reranking to surface the most relevant chunks. Built on FastAPI with async processing, supports multi-format ingestion (PDF, DOCX, images with OCR, URLs), and integrates Google Gemini LLMs with fallback strategies for robustness.

No Package No Dependents
Maintenance 6 / 25
Adoption 3 / 25
Maturity 9 / 25
Community 12 / 25

How are scores calculated?

Stars

3

Forks

1

Language

Python

License

MIT

Last pushed

Nov 02, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/Cerno-AI/Cerno-Insight"

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