LlamaIndex Blog Updates - May 2026
AI · · 1 min read
LlamaIndex Blog Updates (May 2026)
Introducing ParseBench: The First Document Parsing Benchmark for AI Agents
A new open-source benchmark of ~2,000 human-verified enterprise document pages with over 167,000 test rules, evaluating parsers across five dimensions: tables, charts, content faithfulness, semantic formatting, and visual grounding. LlamaParse Agentic scored 84.9% overall - the only parser competitive across all five dimensions.
OCR for KYC: Why Standard Text Extraction Falls Short
Deep dive on why standard OCR fails for Know-Your-Customer compliance, covering worn passports, angled phone photos, and multilingual identity documents. Agentic OCR approach vs traditional.
Mortgage Document Automation: Transforming Loan Processing
How intelligent document processing transforms mortgage lending with production-grade workflows.
Other Recent Posts
- Parsing the Unreadable: How LlamaParse Handles Legal Discovery Documents
- Building a Financial Document Pipeline with LlamaParse
- Building a Financial Due Diligence Agent with LiteParse
- LiteParse passes 4.3K GitHub stars
- Anthropic Opus 4.7 benchmark results (+42% on charts, LlamaParse Agentic still leads at 84.9%)