🌐

Knowledge Base

Build a local chemistry corpus by scraping PubChem, ChEMBL, Wikipedia, USPTO β€” used for RAG retrieval & training datasets
Loading stats…
⬇️ New Scrape Job
⏱ Roughly 0.2-0.5s per result depending on source rate limits.
Duplicates auto-skipped via UNIQUE(source, source_id).
🌱 Quick Seeds
Pre-built scraping queries for common pharma corpora:
πŸ“‹ Recent Scrape Jobs
πŸ” Browse Knowledge Base
Click Search to browse